https://wiki.inria.fr/wikis/rosp/api.php?action=feedcontributions&feedformat=atom&user=Ramon.astudillo rosp - User contributions [en] 2026-08-29T00:58:59Z User contributions MediaWiki 1.32.6 https://wiki.inria.fr/wikis/rosp/index.php?title=List_of_SIG_members&diff=491 List of SIG members 2018-10-25T09:05:50Z <p>Ramon.astudillo: </p> <hr /> <div>139 members as of August 29, 2018:<br /> * Søren Holdt Jensen (Univ. Aalborg)<br /> * Børge Lindberg (Univ. Aalborg)<br /> * Zhen-Hua Tan (Univ. Aalborg)<br /> * Kalle Palomäki (Aalto Univ.)<br /> * Yu Tsao (Academia Sinica)<br /> * Steve Beet (Aculab)<br /> * Gautham Mysore (Adobe)<br /> * Björn Hoffmeister (Amazon)<br /> * Georg Holzman (auphonic.com)<br /> * Dimitris Dimitriadis (AT&amp;T Research)<br /> * Welly Naptali (ATR-Trek <br /> * Nassos Katsamanis (TU Athens)<br /> * Emanuël Habets (Audiolabs Erlangen)<br /> * Keith Ponting (Avaya)<br /> * Daniele Giacobello (Beats by Dr. Dre)<br /> * Armin Sehr (Beuth HS)<br /> * Morgan Nelson (Univ. California, Berkeley)<br /> * Lalan Kumar (IIT Bhubaneswar)<br /> *Jounghoon Beh (Bosch Research and Technology Center)<br /> * Pranav Jawale (IIT Bombay)<br /> * Rajbabu Velmurugan (IIT Bombay)<br /> * Kateřina Žmolíková (Brno University of Technology)<br /> * Mark Gales (Univ. Cambrigde)<br /> * Angel Mario Castro Martínez (Univ. Carl von Ossietzky)<br /> * Bhiksha Raj (Carnegie-Mellon Univ.)<br /> * Rita Singh (Carnegie-Mellon Univ.)<br /> * Richard Stern (Carnegie-Mellon Univ.)<br /> * Climent Nadeu (UP Catalonia)<br /> * Dayana Ribas Gonzalez (CENATAV)<br /> * Nestor Becerra Yoma (Univ. Chile)<br /> * Li Guanjun (Chinese Academy of Sciences)<br /> * Wei Xue (Chinese Academy of Sciences)<br /> * Orpheus Wang (Chinese Academy of Sciences)<br /> * Fabio Tesser (CNR)<br /> * Jahangir Alam (CRIM)<br /> * Gerhard Taljaard (CSIR)<br /> * Petr Mizera (Czech TU)<br /> * Harish-Chandra Dubey (UT Dallas)<br /> * John L. Hansen (UT Dallas)<br /> * Taufiq Hasan (UT Dallas)<br /> * Maryam Najafian (UT Dallas)<br /> * Avni Rajpal (IICT Dhirubhai Ambani)<br /> * Alexey Tarasov (Dublin IT)<br /> * Walter Kellermann (Univ. Erlangen-Nürnberg)<br /> * Roland Maas (Univ. Erlangen-Nürnberg)<br /> * Georg Stemmer (Univ. Erlangen-Nürnberg)<br /> * Alessio Brutti (FBK)<br /> * Cristina Guerrero Flores (FBK)<br /> * Waheed Mohamed (FBK)<br /> * Brigitte Richardson (Ford Motors)<br /> * Thad Hughes (Google)<br /> * Ananya Misra (Google)<br /> * Arun Narayanan (Google)<br /> * Rohit Prabhavalkar (Google)<br /> * Richard Rose (Google)<br /> * Kevin Wilson (Google)<br /> * José Andrés Gonzalez (Univ. Granada)<br /> * Juan Andres Morales Cordovilla (Univ. Granada)<br /> * Juan Azcarreta (Bragi)<br /> * Antonio M. Peinado (Univ. Granada)<br /> * Pejman Mowlaee (Univ. Graz)<br /> * Kwang Myung Jeon (Gwangju IST)<br /> * Martin Heckmann (Honda)<br /> * Ramya Rasipuram (Idiap)<br /> * Alberto Abad (INESC-Id)<br /> * Ramon F. Astudillo (INESC-Id)<br /> * Sunit Sivasankaran (Inria)<br /> * Emmanuel Vincent (Inria)<br /> * David Pearce (Intel)<br /> * Nick Ruiz (Interactions, LLC)<br /> * João Felipe Santos (INRS-EMT)<br /> * Rohit Gowtham (Interactive Intelligence)<br /> * Hynek Hermansky (Johns-Hopkins Univ.)<br /> * Vijayaditya Peddinti (Johns-Hopkins Univ.)<br /> * Sudarsana R. Kadiri (IIIT Hyderabad)<br /> * Rajesh Hegde (IIT Kanpur)<br /> * Karan Nathwani (IIT Kanpur)<br /> * Jort Gemmeke (KU Leuven)<br /> * Hugo van Hamme (KU Leuven)<br /> * Jiri Malek (Univ. Liberec)<br /> * Cong-Thanh Do (LIMSI)<br /> * Arseniy Gorin (LIMSI)<br /> * Rajiv Reddy (LumenVox)<br /> * Alfred Mertins (Univ. Lübeck)<br /> * Ke Chen (Univ. Manchester)<br /> * John Hershey (MERL)<br /> * Jonathan Le Roux (MERL)<br /> * Shinji Watanabe (MERL)<br /> * Joseph P. Campbell (MIT)<br /> * Jim Glass (MIT)<br /> * Najim Dehak (MIT)<br /> * Sonal Joshi (TCS Research &amp; Innovation)<br /> * Björn Schuller (TU Munich)<br /> * Felix Weninger (TU Munich)<br /> * Xiong Xiao (Nanyang Technological Univ.)<br /> * Lou Boves (Univ. Nijmegen)<br /> * Mario Ganzeboom (Univ. Nijmegen)<br /> * Bert Cranen (Univ. Nijmegen)<br /> * Louis ten Bosch (Univ. Nijmegen)<br /> * David van Leeuwen (Univ. Nijmegen)<br /> * Niksa Jakovljevic (Univ. Novi Sad)<br /> * Shuguang Wang (Northwestern Polytechnical Univ., Xi'an)<br /> * Marc Delcroix (NTT)<br /> * Miquel Espi (NTT)<br /> * Keisuke Kinoshita (NTT)<br /> * Dung Tran (NTT)<br /> * Wouter Tirry (NXP Software)<br /> * DeLiang Wang (Ohio State Univ.)<br /> * Zhongqiu Wang (Ohio State Univ.)<br /> * Simon Bozonnet (OnMobile)<br /> * Anil Alexander (Oxford Wave Research)<br /> * Reinhold Häb-Umbach (Univ. Paderborn)<br /> * Quang Nguyen (PowerVoice)<br /> * Ricard Marxer (Univ. of Sheffield)<br /> * Ahmed Hussen Abdelaziz (Ruhr Univ., Bochum)<br /> * Dorothea Kolossa (Ruhr Univ., Bochum)<br /> * Tobias Menne (RWTH University, Aachen)<br /> * Rahil Mahdiyan (Saarland Univ.)<br /> * Hakan Erdogan (Sabanci Univ.)<br /> * Emad Grais (Sabanci Univ.)<br /> * Raghavendra Bilgi (Samsung)<br /> * David Trainor (Sentireal)<br /> * Jon P. Barker (Univ. of Sheffield)<br /> * Kenichi Kumatani (Spansion)<br /> * Jordan Cohen (Spelamode)<br /> * Mahesh Nandwana (SRI International)<br /> * Leslie Smith (Univ. Stirling)<br /> * Mark D. Plumbley (Univ. of Surrey)<br /> * Dominic Ward (Univ. of Surrey)<br /> * Punit Kumar Mohanty (Tata Consultancy Services)<br /> * Gérard Chollet (Télécom ParisTech)<br /> * Otman Chakkor (Univ. Tetuan)<br /> * Ngoc Q.K. Duong (Technicolor)<br /> * George Sterpu (Trinity College Dublin)<br /> * Zhijian Ou (Tsinghua Univ.)<br /> * Antti Hurmalainen (TU Tampere)<br /> * Tuomas Virtanen (TU Tampere)<br /> * Vinay Modi (Voice Pitara Technologies)<br /> * Jidong Tao (Vlingo)<br /> * Raphaël Blouet<br /> * Estelle Delpech</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=List_of_SIG_members&diff=485 List of SIG members 2018-08-29T09:35:30Z <p>Ramon.astudillo: </p> <hr /> <div>138 members as of August 21, 2018:<br /> * Søren Holdt Jensen (Univ. Aalborg)<br /> * Børge Lindberg (Univ. Aalborg)<br /> * Zhen-Hua Tan (Univ. Aalborg)<br /> * Kalle Palomäki (Aalto Univ.)<br /> * Yu Tsao (Academia Sinica)<br /> * Steve Beet (Aculab)<br /> * Gautham Mysore (Adobe)<br /> * Björn Hoffmeister (Amazon)<br /> * Georg Holzman (auphonic.com)<br /> * Dimitris Dimitriadis (AT&amp;T Research)<br /> * Welly Naptali (ATR-Trek <br /> * Nassos Katsamanis (TU Athens)<br /> * Emanuël Habets (Audiolabs Erlangen)<br /> * Keith Ponting (Avaya)<br /> * Daniele Giacobello (Beats by Dr. Dre)<br /> * Armin Sehr (Beuth HS)<br /> * Morgan Nelson (Univ. California, Berkeley)<br /> * Lalan Kumar (IIT Bhubaneswar)<br /> *Jounghoon Beh (Bosch Research and Technology Center)<br /> * Pranav Jawale (IIT Bombay)<br /> * Rajbabu Velmurugan (IIT Bombay)<br /> * Kateřina Žmolíková (Brno University of Technology)<br /> * Mark Gales (Univ. Cambrigde)<br /> * Angel Mario Castro Martínez (Univ. Carl von Ossietzky)<br /> * Bhiksha Raj (Carnegie-Mellon Univ.)<br /> * Rita Singh (Carnegie-Mellon Univ.)<br /> * Richard Stern (Carnegie-Mellon Univ.)<br /> * Climent Nadeu (UP Catalonia)<br /> * Dayana Ribas Gonzalez (CENATAV)<br /> * Nestor Becerra Yoma (Univ. Chile)<br /> * Li Guanjun (Chinese Academy of Sciences)<br /> * Wei Xue (Chinese Academy of Sciences)<br /> * Orpheus Wang (Chinese Academy of Sciences)<br /> * Fabio Tesser (CNR)<br /> * Jahangir Alam (CRIM)<br /> * Gerhard Taljaard (CSIR)<br /> * Petr Mizera (Czech TU)<br /> * Harish-Chandra Dubey (UT Dallas)<br /> * John L. Hansen (UT Dallas)<br /> * Taufiq Hasan (UT Dallas)<br /> * Maryam Najafian (UT Dallas)<br /> * Avni Rajpal (IICT Dhirubhai Ambani)<br /> * Alexey Tarasov (Dublin IT)<br /> * Walter Kellermann (Univ. Erlangen-Nürnberg)<br /> * Roland Maas (Univ. Erlangen-Nürnberg)<br /> * Georg Stemmer (Univ. Erlangen-Nürnberg)<br /> * Alessio Brutti (FBK)<br /> * Cristina Guerrero Flores (FBK)<br /> * Waheed Mohamed (FBK)<br /> * Brigitte Richardson (Ford Motors)<br /> * Thad Hughes (Google)<br /> * Ananya Misra (Google)<br /> * Arun Narayanan (Google)<br /> * Rohit Prabhavalkar (Google)<br /> * Richard Rose (Google)<br /> * Kevin Wilson (Google)<br /> * José Andrés Gonzalez (Univ. Granada)<br /> * Juan Andres Morales Cordovilla (Univ. Granada)<br /> * Juan Azcarreta (Bragi)<br /> * Antonio M. Peinado (Univ. Granada)<br /> * Pejman Mowlaee (Univ. Graz)<br /> * Kwang Myung Jeon (Gwangju IST)<br /> * Martin Heckmann (Honda)<br /> * Ramya Rasipuram (Idiap)<br /> * Alberto Abad (INESC-Id)<br /> * Ramon F. Astudillo (INESC-Id)<br /> * Sunit Sivasankaran (Inria)<br /> * Emmanuel Vincent (Inria)<br /> * David Pearce (Intel)<br /> * Nick Ruiz (Interactions, LLC)<br /> * João Felipe Santos (INRS-EMT)<br /> * Rohit Gowtham (Interactive Intelligence)<br /> * Hynek Hermansky (Johns-Hopkins Univ.)<br /> * Vijayaditya Peddinti (Johns-Hopkins Univ.)<br /> * Sudarsana R. Kadiri (IIIT Hyderabad)<br /> * Rajesh Hegde (IIT Kanpur)<br /> * Karan Nathwani (IIT Kanpur)<br /> * Jort Gemmeke (KU Leuven)<br /> * Hugo van Hamme (KU Leuven)<br /> * Jiri Malek (Univ. Liberec)<br /> * Cong-Thanh Do (LIMSI)<br /> * Arseniy Gorin (LIMSI)<br /> * Rajiv Reddy (LumenVox)<br /> * Alfred Mertins (Univ. Lübeck)<br /> * Ke Chen (Univ. Manchester)<br /> * John Hershey (MERL)<br /> * Jonathan Le Roux (MERL)<br /> * Shinji Watanabe (MERL)<br /> * Joseph P. Campbell (MIT)<br /> * Jim Glass (MIT)<br /> * Najim Dehak (MIT)<br /> * Björn Schuller (TU Munich)<br /> * Felix Weninger (TU Munich)<br /> * Xiong Xiao (Nanyang Technological Univ.)<br /> * Lou Boves (Univ. Nijmegen)<br /> * Mario Ganzeboom (Univ. Nijmegen)<br /> * Bert Cranen (Univ. Nijmegen)<br /> * Louis ten Bosch (Univ. Nijmegen)<br /> * David van Leeuwen (Univ. Nijmegen)<br /> * Niksa Jakovljevic (Univ. Novi Sad)<br /> * Shuguang Wang (Northwestern Polytechnical Univ., Xi'an)<br /> * Marc Delcroix (NTT)<br /> * Miquel Espi (NTT)<br /> * Keisuke Kinoshita (NTT)<br /> * Dung Tran (NTT)<br /> * Wouter Tirry (NXP Software)<br /> * DeLiang Wang (Ohio State Univ.)<br /> * Zhongqiu Wang (Ohio State Univ.)<br /> * Simon Bozonnet (OnMobile)<br /> * Anil Alexander (Oxford Wave Research)<br /> * Reinhold Häb-Umbach (Univ. Paderborn)<br /> * Quang Nguyen (PowerVoice)<br /> * Ricard Marxer (Univ. of Sheffield)<br /> * Ahmed Hussen Abdelaziz (Ruhr Univ., Bochum)<br /> * Dorothea Kolossa (Ruhr Univ., Bochum)<br /> * Tobias Menne (RWTH University, Aachen)<br /> * Rahil Mahdiyan (Saarland Univ.)<br /> * Hakan Erdogan (Sabanci Univ.)<br /> * Emad Grais (Sabanci Univ.)<br /> * Raghavendra Bilgi (Samsung)<br /> * David Trainor (Sentireal)<br /> * Jon P. Barker (Univ. of Sheffield)<br /> * Kenichi Kumatani (Spansion)<br /> * Jordan Cohen (Spelamode)<br /> * Mahesh Nandwana (SRI International)<br /> * Leslie Smith (Univ. Stirling)<br /> * Mark D. Plumbley (Univ. of Surrey)<br /> * Dominic Ward (Univ. of Surrey)<br /> * Punit Kumar Mohanty (Tata Consultancy Services)<br /> * Gérard Chollet (Télécom ParisTech)<br /> * Otman Chakkor (Univ. Tetuan)<br /> * Ngoc Q.K. Duong (Technicolor)<br /> * George Sterpu (Trinity College Dublin)<br /> * Zhijian Ou (Tsinghua Univ.)<br /> * Antti Hurmalainen (TU Tampere)<br /> * Tuomas Virtanen (TU Tampere)<br /> * Vinay Modi (Voice Pitara Technologies)<br /> * Jidong Tao (Vlingo)<br /> * Raphaël Blouet<br /> * Estelle Delpech</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=List_of_SIG_members&diff=481 List of SIG members 2018-06-25T16:21:08Z <p>Ramon.astudillo: </p> <hr /> <div>139 members as of January 18, 2018:<br /> * Søren Holdt Jensen (Univ. Aalborg)<br /> * Børge Lindberg (Univ. Aalborg)<br /> * Zhen-Hua Tan (Univ. Aalborg)<br /> * Kalle Palomäki (Aalto Univ.)<br /> * Steve Beet (Aculab)<br /> * Gautham Mysore (Adobe)<br /> * Björn Hoffmeister (Amazon)<br /> * Georg Holzman (auphonic.com)<br /> * Dimitris Dimitriadis (AT&amp;T Research)<br /> * Welly Naptali (ATR-Trek <br /> * Nassos Katsamanis (TU Athens)<br /> * Emanuël Habets (Audiolabs Erlangen)<br /> * Keith Ponting (Avaya)<br /> * Daniele Giacobello (Beats by Dr. Dre)<br /> * Armin Sehr (Beuth HS)<br /> * Morgan Nelson (Univ. California, Berkeley)<br /> * Lalan Kumar (IIT Bhubaneswar)<br /> *Jounghoon Beh (Bosch Research and Technology Center)<br /> * Pranav Jawale (IIT Bombay)<br /> * Rajbabu Velmurugan (IIT Bombay)<br /> * Kateřina Žmolíková (Brno University of Technology)<br /> * Mark Gales (Univ. Cambrigde)<br /> * Angel Mario Castro Martínez (Univ. Carl von Ossietzky)<br /> * Bhiksha Raj (Carnegie-Mellon Univ.)<br /> * Rita Singh (Carnegie-Mellon Univ.)<br /> * Richard Stern (Carnegie-Mellon Univ.)<br /> * Climent Nadeu (UP Catalonia)<br /> * Dayana Ribas Gonzalez (CENATAV)<br /> * Nestor Becerra Yoma (Univ. Chile)<br /> * Wei Xue (Chinese Academy of Sciences)<br /> * Orpheus Wang (Chinese Academy of Sciences)<br /> * Fabio Tesser (CNR)<br /> * Jahangir Alam (CRIM)<br /> * Gerhard Taljaard (CSIR)<br /> * Petr Mizera (Czech TU)<br /> * Harish-Chandra Dubey (UT Dallas)<br /> * John L. Hansen (UT Dallas)<br /> * Taufiq Hasan (UT Dallas)<br /> * Maryam Najafian (UT Dallas)<br /> * Avni Rajpal (IICT Dhirubhai Ambani)<br /> * Alexey Tarasov (Dublin IT)<br /> * Walter Kellermann (Univ. Erlangen-Nürnberg)<br /> * Roland Maas (Univ. Erlangen-Nürnberg)<br /> * Georg Stemmer (Univ. Erlangen-Nürnberg)<br /> * Alessio Brutti (FBK)<br /> * Cristina Guerrero Flores (FBK)<br /> * Waheed Mohamed (FBK)<br /> * Brigitte Richardson (Ford Motors)<br /> * Thad Hughes (Google)<br /> * Ananya Misra (Google)<br /> * Arun Narayanan (Google)<br /> * Rohit Prabhavalkar (Google)<br /> * Richard Rose (Google)<br /> * Kevin Wilson (Google)<br /> * José Andrés Gonzalez (Univ. Granada)<br /> * Juan Andres Morales Cordovilla (Univ. Granada)<br /> * Juan Azcarreta (Bragi)<br /> * Antonio M. Peinado (Univ. Granada)<br /> * Pejman Mowlaee (Univ. Graz)<br /> * Kwang Myung Jeon (Gwangju IST)<br /> * Martin Heckmann (Honda)<br /> * Afsaneh Asaei (Idiap)<br /> * Ramya Rasipuram (Idiap)<br /> * Alberto Abad (INESC-Id)<br /> * Ramon F. Astudillo (INESC-Id)<br /> * Sunit Sivasankaran (Inria)<br /> * Emmanuel Vincent (Inria)<br /> * David Pearce (Intel)<br /> * Nick Ruiz (Interactions, LLC)<br /> * João Felipe Santos (INRS-EMT)<br /> * Rohit Gowtham (Interactive Intelligence)<br /> * Hynek Hermansky (Johns-Hopkins Univ.)<br /> * Vijayaditya Peddinti (Johns-Hopkins Univ.)<br /> * Sudarsana R. Kadiri (IIIT Hyderabad)<br /> * Rajesh Hegde (IIT Kanpur)<br /> * Karan Nathwani (IIT Kanpur)<br /> * Jort Gemmeke (KU Leuven)<br /> * Hugo van Hamme (KU Leuven)<br /> * Jiri Malek (Univ. Liberec)<br /> * Cong-Thanh Do (LIMSI)<br /> * Arseniy Gorin (LIMSI)<br /> * Rajiv Reddy (LumenVox)<br /> * Alfred Mertins (Univ. Lübeck)<br /> * Florian Müller (Univ. Lübeck)<br /> * Ke Chen (Univ. Manchester)<br /> * John Hershey (MERL)<br /> * Jonathan Le Roux (MERL)<br /> * Shinji Watanabe (MERL)<br /> * Mike Seltzer (Microsoft)<br /> * Joseph P. Campbell (MIT)<br /> * Jim Glass (MIT)<br /> * Najim Dehak (MIT)<br /> * Björn Schuller (TU Munich)<br /> * Felix Weninger (TU Munich)<br /> * Xiong Xiao (Nanyang Technological Univ.)<br /> * Lou Boves (Univ. Nijmegen)<br /> * Mario Ganzeboom (Univ. Nijmegen)<br /> * Bert Cranen (Univ. Nijmegen)<br /> * Louis ten Bosch (Univ. Nijmegen)<br /> * David van Leeuwen (Univ. Nijmegen)<br /> * Niksa Jakovljevic (Univ. Novi Sad)<br /> * Shuguang Wang (Northwestern Polytechnical Univ., Xi'an)<br /> * Marc Delcroix (NTT)<br /> * Miquel Espi (NTT)<br /> * Keisuke Kinoshita (NTT)<br /> * Dung Tran (NTT)<br /> * Wouter Tirry (NXP Software)<br /> * DeLiang Wang (Ohio State Univ.)<br /> * Zhongqiu Wang (Ohio State Univ.)<br /> * Simon Bozonnet (OnMobile)<br /> * Anil Alexander (Oxford Wave Research)<br /> * Reinhold Häb-Umbach (Univ. Paderborn)<br /> * Quang Nguyen (PowerVoice)<br /> * Ricard Marxer (Univ. of Sheffield)<br /> * Ahmed Hussen Abdelaziz (Ruhr Univ., Bochum)<br /> * Dorothea Kolossa (Ruhr Univ., Bochum)<br /> * Tobias Menne (RWTH University, Aachen)<br /> * Rahil Mahdiyan (Saarland Univ.)<br /> * Hakan Erdogan (Sabanci Univ.)<br /> * Emad Grais (Sabanci Univ.)<br /> * Raghavendra Bilgi (Samsung)<br /> * David Trainor (Sentireal)<br /> * Jon P. Barker (Univ. of Sheffield)<br /> * Kenichi Kumatani (Spansion)<br /> * Jordan Cohen (Spelamode)<br /> * Mahesh Nandwana (SRI International)<br /> * Leslie Smith (Univ. Stirling)<br /> * Mark D. Plumbley (Univ. of Surrey)<br /> * Dominic Ward (Univ. of Surrey)<br /> * Punit Kumar Mohanty (Tata Consultancy Services)<br /> * Gérard Chollet (Télécom ParisTech)<br /> * Otman Chakkor (Univ. Tetuan)<br /> * Ngoc Q.K. Duong (Technicolor)<br /> * George Sterpu (Trinity College Dublin)<br /> * Zhijian Ou (Tsinghua Univ.)<br /> * Antti Hurmalainen (TU Tampere)<br /> * Tuomas Virtanen (TU Tampere)<br /> * Vinay Modi (Voice Pitara Technologies)<br /> * Jidong Tao (Vlingo)<br /> * Raphaël Blouet<br /> * Estelle Delpech</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=List_of_SIG_members&diff=478 List of SIG members 2018-01-15T17:14:56Z <p>Ramon.astudillo: </p> <hr /> <div>137 members as of October 24, 2017:<br /> * Søren Holdt Jensen (Univ. Aalborg)<br /> * Børge Lindberg (Univ. Aalborg)<br /> * Zhen-Hua Tan (Univ. Aalborg)<br /> * Kalle Palomäki (Aalto Univ.)<br /> * Steve Beet (Aculab)<br /> * Gautham Mysore (Adobe)<br /> * Björn Hoffmeister (Amazon)<br /> * Georg Holzman (auphonic.com)<br /> * Dimitris Dimitriadis (AT&amp;T Research)<br /> * Welly Naptali (ATR-Trek <br /> * Nassos Katsamanis (TU Athens)<br /> * Emanuël Habets (Audiolabs Erlangen)<br /> * Keith Ponting (Avaya)<br /> * Daniele Giacobello (Beats by Dr. Dre)<br /> * Armin Sehr (Beuth HS)<br /> * Morgan Nelson (Univ. California, Berkeley)<br /> * Lalan Kumar (IIT Bhubaneswar)<br /> *Jounghoon Beh (Bosch Research and Technology Center)<br /> * Pranav Jawale (IIT Bombay)<br /> * Rajbabu Velmurugan (IIT Bombay)<br /> * Kateřina Žmolíková (Brno University of Technology)<br /> * Mark Gales (Univ. Cambrigde)<br /> * Angel Mario Castro Martínez (Univ. Carl von Ossietzky)<br /> * Bhiksha Raj (Carnegie-Mellon Univ.)<br /> * Rita Singh (Carnegie-Mellon Univ.)<br /> * Richard Stern (Carnegie-Mellon Univ.)<br /> * Climent Nadeu (UP Catalonia)<br /> * Dayana Ribas Gonzalez (CENATAV)<br /> * Nestor Becerra Yoma (Univ. Chile)<br /> * Wei Xue (Chinese Academy of Sciences)<br /> * Orpheus Wang (Chinese Academy of Sciences)<br /> * Fabio Tesser (CNR)<br /> * Jahangir Alam (CRIM)<br /> * Gerhard Taljaard (CSIR)<br /> * Petr Mizera (Czech TU)<br /> * Harish-Chandra Dubey (UT Dallas)<br /> * John L. Hansen (UT Dallas)<br /> * Taufiq Hasan (UT Dallas)<br /> * Maryam Najafian (UT Dallas)<br /> * Avni Rajpal (IICT Dhirubhai Ambani)<br /> * Alexey Tarasov (Dublin IT)<br /> * Walter Kellermann (Univ. Erlangen-Nürnberg)<br /> * Roland Maas (Univ. Erlangen-Nürnberg)<br /> * Georg Stemmer (Univ. Erlangen-Nürnberg)<br /> * Alessio Brutti (FBK)<br /> * Cristina Guerrero Flores (FBK)<br /> * Waheed Mohamed (FBK)<br /> * Brigitte Richardson (Ford Motors)<br /> * Thad Hughes (Google)<br /> * Ananya Misra (Google)<br /> * Arun Narayanan (Google)<br /> * Rohit Prabhavalkar (Google)<br /> * Richard Rose (Google)<br /> * Kevin Wilson (Google)<br /> * José Andrés Gonzalez (Univ. Granada)<br /> * Juan Andres Morales Cordovilla (Univ. Granada)<br /> * Antonio M. Peinado (Univ. Granada)<br /> * Pejman Mowlaee (Univ. Graz)<br /> * Kwang Myung Jeon (Gwangju IST)<br /> * Martin Heckmann (Honda)<br /> * Afsaneh Asaei (Idiap)<br /> * Ramya Rasipuram (Idiap)<br /> * Alberto Abad (INESC-Id)<br /> * Ramon F. Astudillo (INESC-Id)<br /> * Sunit Sivasankaran (Inria)<br /> * Emmanuel Vincent (Inria)<br /> * David Pearce (Intel)<br /> * Nick Ruiz (Interactions, LLC)<br /> * João Felipe Santos (INRS-EMT)<br /> * Rohit Gowtham (Interactive Intelligence)<br /> * Hynek Hermansky (Johns-Hopkins Univ.)<br /> * Vijayaditya Peddinti (Johns-Hopkins Univ.)<br /> * Sudarsana R. Kadiri (IIIT Hyderabad)<br /> * Rajesh Hegde (IIT Kanpur)<br /> * Karan Nathwani (IIT Kanpur)<br /> * Jort Gemmeke (KU Leuven)<br /> * Hugo van Hamme (KU Leuven)<br /> * Jiri Malek (Univ. Liberec)<br /> * Cong-Thanh Do (LIMSI)<br /> * Arseniy Gorin (LIMSI)<br /> * Rajiv Reddy (LumenVox)<br /> * Alfred Mertins (Univ. Lübeck)<br /> * Florian Müller (Univ. Lübeck)<br /> * Ke Chen (Univ. Manchester)<br /> * John Hershey (MERL)<br /> * Jonathan Le Roux (MERL)<br /> * Shinji Watanabe (MERL)<br /> * Mike Seltzer (Microsoft)<br /> * Joseph P. Campbell (MIT)<br /> * Jim Glass (MIT)<br /> * Najim Dehak (MIT)<br /> * Björn Schuller (TU Munich)<br /> * Felix Weninger (TU Munich)<br /> * Xiong Xiao (Nanyang Technological Univ.)<br /> * Lou Boves (Univ. Nijmegen)<br /> * Mario Ganzeboom (Univ. Nijmegen)<br /> * Bert Cranen (Univ. Nijmegen)<br /> * Louis ten Bosch (Univ. Nijmegen)<br /> * David van Leeuwen (Univ. Nijmegen)<br /> * Niksa Jakovljevic (Univ. Novi Sad)<br /> * Shuguang Wang (Northwestern Polytechnical Univ., Xi'an)<br /> * Marc Delcroix (NTT)<br /> * Miquel Espi (NTT)<br /> * Keisuke Kinoshita (NTT)<br /> * Dung Tran (NTT)<br /> * Wouter Tirry (NXP Software)<br /> * DeLiang Wang (Ohio State Univ.)<br /> * Zhongqiu Wang (Ohio State Univ.)<br /> * Simon Bozonnet (OnMobile)<br /> * Anil Alexander (Oxford Wave Research)<br /> * Reinhold Häb-Umbach (Univ. Paderborn)<br /> * Quang Nguyen (PowerVoice)<br /> * Ricard Marxer (Univ. of Sheffield)<br /> * Ahmed Hussen Abdelaziz (Ruhr Univ., Bochum)<br /> * Dorothea Kolossa (Ruhr Univ., Bochum)<br /> * Tobias Menne (RWTH University, Aachen)<br /> * Rahil Mahdiyan (Saarland Univ.)<br /> * Hakan Erdogan (Sabanci Univ.)<br /> * Emad Grais (Sabanci Univ.)<br /> * Raghavendra Bilgi (Samsung)<br /> * David Trainor (Sentireal)<br /> * Jon P. Barker (Univ. of Sheffield)<br /> * Kenichi Kumatani (Spansion)<br /> * Jordan Cohen (Spelamode)<br /> * Mahesh Nandwana (SRI International)<br /> * Leslie Smith (Univ. Stirling)<br /> * Dominic Ward (Univ. of Surrey)<br /> * Punit Kumar Mohanty (Tata Consultancy Services)<br /> * Gérard Chollet (Télécom ParisTech)<br /> * Otman Chakkor (Univ. Tetuan)<br /> * Ngoc Q.K. Duong (Technicolor)<br /> * George Sterpu (Trinity College Dublin)<br /> * Zhijian Ou (Tsinghua Univ.)<br /> * Antti Hurmalainen (TU Tampere)<br /> * Tuomas Virtanen (TU Tampere)<br /> * Vinay Modi (Voice Pitara Technologies)<br /> * Jidong Tao (Vlingo)<br /> * Raphaël Blouet<br /> * Estelle Delpech</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=List_of_SIG_members&diff=468 List of SIG members 2017-02-21T22:00:59Z <p>Ramon.astudillo: </p> <hr /> <div>133 members as of January 6, 2017:<br /> * Søren Holdt Jensen (Univ. Aalborg)<br /> * Børge Lindberg (Univ. Aalborg)<br /> * Zhen-Hua Tan (Univ. Aalborg)<br /> * Kalle Palomäki (Aalto Univ.)<br /> * Steve Beet (Aculab)<br /> * Gautham Mysore (Adobe)<br /> * Björn Hoffmeister (Amazon)<br /> * Georg Holzman (auphonic.com)<br /> * Dimitris Dimitriadis (AT&amp;T Research)<br /> * Welly Naptali (ATR-Trek <br /> * Nassos Katsamanis (TU Athens)<br /> * David Pearce (Audience)<br /> * Keith Ponting (Avaya)<br /> * Daniele Giacobello (Beats by Dr. Dre)<br /> * Armin Sehr (Beuth HS)<br /> * Morgan Nelson (Univ. California, Berkeley)<br /> * Lalan Kumar (IIT Bhubaneswar)<br /> * Pranav Jawale (IIT Bombay)<br /> * Rajbabu Velmurugan (IIT Bombay)<br /> * Kateřina Žmolíková (Brno University of Technology)<br /> * Mark Gales (Univ. Cambrigde)<br /> * Angel Mario Castro Martínez (Univ. Carl von Ossietzky)<br /> * Bhiksha Raj (Carnegie-Mellon Univ.)<br /> * Rita Singh (Carnegie-Mellon Univ.)<br /> * Richard Stern (Carnegie-Mellon Univ.)<br /> * Climent Nadeu (UP Catalonia)<br /> * Dayana Ribas Gonzalez (CENATAV)<br /> * Nestor Becerra Yoma (Univ. Chile)<br /> * Wei Xue (Chinese Academy of Sciences)<br /> * Orpheus Wang (Chinese Academy of Sciences)<br /> * Fabio Tesser (CNR)<br /> * Jahangir Alam (CRIM)<br /> * Gerhard Taljaard (CSIR)<br /> * Petr Mizera (Czech TU)<br /> * Harish-Chandra Dubey (UT Dallas)<br /> * John L. Hansen (UT Dallas)<br /> * Taufiq Hasan (UT Dallas)<br /> * Maryam Najafian (UT Dallas)<br /> * Avni Rajpal (IICT Dhirubhai Ambani)<br /> * Alexey Tarasov (Dublin IT)<br /> * Emanuël Habets (Univ. Erlangen-Nürnberg)<br /> * Walter Kellermann (Univ. Erlangen-Nürnberg)<br /> * Roland Maas (Univ. Erlangen-Nürnberg)<br /> * Georg Stemmer (Univ. Erlangen-Nürnberg)<br /> * Alessio Brutti (FBK)<br /> * Cristina Guerrero Flores (FBK)<br /> * Waheed Mohamed (FBK)<br /> * Brigitte Richardson (Ford Motors)<br /> * Thad Hughes (Google)<br /> * Ananya Misra (Google)<br /> * Arun Narayanan (Google)<br /> * Rohit Prabhavalkar (Google)<br /> * Richard Rose (Google)<br /> * Kevin Wilson (Google)<br /> * José Andrés Gonzalez (Univ. Granada)<br /> * Juan Andres Morales Cordovilla (Univ. Granada)<br /> * Antonio M. Peinado (Univ. Granada)<br /> * Pejman Mowlaee (Univ. Graz)<br /> * Kwang Myung Jeon (Gwangju IST)<br /> * Martin Heckmann (Honda)<br /> * Afsaneh Asaei (Idiap)<br /> * Ramya Rasipuram (Idiap)<br /> * Alberto Abad (INESC-Id)<br /> * Ramon F. Astudillo (INESC-Id)<br /> * Sunit Sivasankaran (Inria)<br /> * Emmanuel Vincent (Inria)<br /> * Nick Ruiz (Interactions, LLC)<br /> * João Felipe Santos (INRS-EMT)<br /> * Rohit Gowtham (Interactive Intelligence)<br /> * Hynek Hermansky (Johns-Hopkins Univ.)<br /> * Vijayaditya Peddinti (Johns-Hopkins Univ.)<br /> * Rajesh Hegde (IIT Kanpur)<br /> * Karan Nathwani (IIT Kanpur)<br /> * Jort Gemmeke (KU Leuven)<br /> * Hugo van Hamme (KU Leuven)<br /> * Jiri Malek (Univ. Liberec)<br /> * Cong-Thanh Do (LIMSI)<br /> * Arseniy Gorin (LIMSI)<br /> * Rajiv Reddy (LumenVox)<br /> * Alfred Mertins (Univ. Lübeck)<br /> * Florian Müller (Univ. Lübeck)<br /> * Ke Chen (Univ. Manchester)<br /> * John Hershey (MERL)<br /> * Jonathan Le Roux (MERL)<br /> * Shinji Watanabe (MERL)<br /> * Mike Seltzer (Microsoft)<br /> * Joseph P. Campbell (MIT)<br /> * Jim Glass (MIT)<br /> * Najim Dehak (MIT)<br /> * Björn Schuller (TU Munich)<br /> * Felix Weninger (TU Munich)<br /> * Xiong Xiao (Nanyang Technological Univ.)<br /> * Lou Boves (Univ. Nijmegen)<br /> * Mario Ganzeboom (Univ. Nijmegen)<br /> * Bert Cranen (Univ. Nijmegen)<br /> * Louis ten Bosch (Univ. Nijmegen)<br /> * David van Leeuwen (Univ. Nijmegen)<br /> * Niksa Jakovljevic (Univ. Novi Sad)<br /> * Shuguang Wang (Northwestern Polytechnical Univ., Xi'an)<br /> * Marc Delcroix (NTT)<br /> * Miquel Espi (NTT)<br /> * Keisuke Kinoshita (NTT)<br /> * Dung Tran (NTT)<br /> * Cyril Guillaumé (NXP Software)<br /> * Wouter Tirry (NXP Software)<br /> * DeLiang Wang (Ohio State Univ.)<br /> * Zhongqiu Wang (Ohio State Univ.)<br /> * Simon Bozonnet (OnMobile)<br /> * Anil Alexander (Oxford Wave Research)<br /> * Reinhold Häb-Umbach (Univ. Paderborn)<br /> * Ricard Marxer (Univ. of Sheffield)<br /> * Ahmed Hussen Abdelaziz (Ruhr Univ., Bochum)<br /> * Dorothea Kolossa (Ruhr Univ., Bochum)<br /> * Tobias Menne (RWTH University, Aachen)<br /> * Rahil Mahdiyan (Saarland Univ.)<br /> * Hakan Erdogan (Sabanci Univ.)<br /> * Emad Grais (Sabanci Univ.)<br /> * Raghavendra Bilgi (Samsung)<br /> * David Trainor (Sentireal)<br /> * Jon P. Barker (Univ. of Sheffield)<br /> * Kenichi Kumatani (Spansion)<br /> * Jordan Cohen (Spelamode)<br /> * Leslie Smith (Univ. Stirling)<br /> * Punit Kumar Mohanty (Tata Consultancy Services)<br /> * Gérard Chollet (Télécom ParisTech)<br /> * Otman Chakkor (Univ. Tetuan)<br /> * Ngoc Q.K. Duong (Technicolor)<br /> * George Sterpu (Trinity College Dublin)<br /> * Zhijian Ou (Tsinghua Univ.)<br /> * Antti Hurmalainen (TU Tampere)<br /> * Tuomas Virtanen (TU Tampere)<br /> * Vinay Modi (Voice Pitara Technologies)<br /> * Jidong Tao (Vlingo)<br /> * Raphaël Blouet<br /> * Estelle Delpech</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=List_of_SIG_members&diff=464 List of SIG members 2017-01-08T10:25:45Z <p>Ramon.astudillo: </p> <hr /> <div>132 members as of November 24, 2016:<br /> * Søren Holdt Jensen (Univ. Aalborg)<br /> * Børge Lindberg (Univ. Aalborg)<br /> * Zhen-Hua Tan (Univ. Aalborg)<br /> * Kalle Palomäki (Aalto Univ.)<br /> * Steve Beet (Aculab)<br /> * Gautham Mysore (Adobe)<br /> * Björn Hoffmeister (Amazon)<br /> * Dimitris Dimitriadis (AT&amp;T Research)<br /> * Welly Naptali (ATR-Trek <br /> * Nassos Katsamanis (TU Athens)<br /> * David Pearce (Audience)<br /> * Keith Ponting (Avaya)<br /> * Daniele Giacobello (Beats by Dr. Dre)<br /> * Armin Sehr (Beuth HS)<br /> * Morgan Nelson (Univ. California, Berkeley)<br /> * Lalan Kumar (IIT Bhubaneswar)<br /> * Pranav Jawale (IIT Bombay)<br /> * Rajbabu Velmurugan (IIT Bombay)<br /> * Kateřina Žmolíková (Brno University of Technology)<br /> * Mark Gales (Univ. Cambrigde)<br /> * Angel Mario Castro Martínez (Univ. Carl von Ossietzky)<br /> * Bhiksha Raj (Carnegie-Mellon Univ.)<br /> * Rita Singh (Carnegie-Mellon Univ.)<br /> * Richard Stern (Carnegie-Mellon Univ.)<br /> * Climent Nadeu (UP Catalonia)<br /> * Dayana Ribas Gonzalez (CENATAV)<br /> * Nestor Becerra Yoma (Univ. Chile)<br /> * Wei Xue (Chinese Academy of Sciences)<br /> * Orpheus Wang (Chinese Academy of Sciences)<br /> * Fabio Tesser (CNR)<br /> * Jahangir Alam (CRIM)<br /> * Gerhard Taljaard (CSIR)<br /> * Petr Mizera (Czech TU)<br /> * Harish-Chandra Dubey (UT Dallas)<br /> * John L. Hansen (UT Dallas)<br /> * Taufiq Hasan (UT Dallas)<br /> * Maryam Najafian (UT Dallas)<br /> * Avni Rajpal (IICT Dhirubhai Ambani)<br /> * Alexey Tarasov (Dublin IT)<br /> * Emanuël Habets (Univ. Erlangen-Nürnberg)<br /> * Walter Kellermann (Univ. Erlangen-Nürnberg)<br /> * Roland Maas (Univ. Erlangen-Nürnberg)<br /> * Georg Stemmer (Univ. Erlangen-Nürnberg)<br /> * Alessio Brutti (FBK)<br /> * Cristina Guerrero Flores (FBK)<br /> * Waheed Mohamed (FBK)<br /> * Brigitte Richardson (Ford Motors)<br /> * Thad Hughes (Google)<br /> * Ananya Misra (Google)<br /> * Arun Narayanan (Google)<br /> * Rohit Prabhavalkar (Google)<br /> * Kevin Wilson (Google)<br /> * José Andrés Gonzalez (Univ. Granada)<br /> * Juan Andres Morales Cordovilla (Univ. Granada)<br /> * Antonio M. Peinado (Univ. Granada)<br /> * Pejman Mowlaee (Univ. Graz)<br /> * Kwang Myung Jeon (Gwangju IST)<br /> * Martin Heckmann (Honda)<br /> * Afsaneh Asaei (Idiap)<br /> * Ramya Rasipuram (Idiap)<br /> * Alberto Abad (INESC-Id)<br /> * Ramon F. Astudillo (INESC-Id)<br /> * Sunit Sivasankaran (Inria)<br /> * Emmanuel Vincent (Inria)<br /> * Nick Ruiz (Interactions, LLC)<br /> * João Felipe Santos (INRS-EMT)<br /> * Rohit Gowtham (Interactive Intelligence)<br /> * Hynek Hermansky (Johns-Hopkins Univ.)<br /> * Vijayaditya Peddinti (Johns-Hopkins Univ.)<br /> * Rajesh Hegde (IIT Kanpur)<br /> * Karan Nathwani (IIT Kanpur)<br /> * Jort Gemmeke (KU Leuven)<br /> * Hugo van Hamme (KU Leuven)<br /> * Jiri Malek (Univ. Liberec)<br /> * Cong-Thanh Do (LIMSI)<br /> * Arseniy Gorin (LIMSI)<br /> * Rajiv Reddy (LumenVox)<br /> * Alfred Mertins (Univ. Lübeck)<br /> * Florian Müller (Univ. Lübeck)<br /> * Ke Chen (Univ. Manchester)<br /> * Richard Rose (McGill Univ.)<br /> * John Hershey (MERL)<br /> * Jonathan Le Roux (MERL)<br /> * Shinji Watanabe (MERL)<br /> * Mike Seltzer (Microsoft)<br /> * Joseph P. Campbell (MIT)<br /> * Jim Glass (MIT)<br /> * Najim Dehak (MIT)<br /> * Björn Schuller (TU Munich)<br /> * Felix Weninger (TU Munich)<br /> * Xiong Xiao (Nanyang Technological Univ.)<br /> * Lou Boves (Univ. Nijmegen)<br /> * Mario Ganzeboom (Univ. Nijmegen)<br /> * Bert Cranen (Univ. Nijmegen)<br /> * Louis ten Bosch (Univ. Nijmegen)<br /> * David van Leeuwen (Univ. Nijmegen)<br /> * Niksa Jakovljevic (Univ. Novi Sad)<br /> * Shuguang Wang (Northwestern Polytechnical Univ., Xi'an)<br /> * Marc Delcroix (NTT)<br /> * Miquel Espi (NTT)<br /> * Keisuke Kinoshita (NTT)<br /> * Dung Tran (NTT)<br /> * Cyril Guillaumé (NXP Software)<br /> * Wouter Tirry (NXP Software)<br /> * DeLiang Wang (Ohio State Univ.)<br /> * Zhongqiu Wang (Ohio State Univ.)<br /> * Simon Bozonnet (OnMobile)<br /> * Anil Alexander (Oxford Wave Research)<br /> * Reinhold Häb-Umbach (Univ. Paderborn)<br /> * Ricard Marxer (Univ. of Sheffield)<br /> * Ahmed Hussen Abdelaziz (Ruhr Univ., Bochum)<br /> * Dorothea Kolossa (Ruhr Univ., Bochum)<br /> * Tobias Menne (RWTH University, Aachen)<br /> * Rahil Mahdiyan (Saarland Univ.)<br /> * Hakan Erdogan (Sabanci Univ.)<br /> * Emad Grais (Sabanci Univ.)<br /> * Raghavendra Bilgi (Samsung)<br /> * David Trainor (Sentireal)<br /> * Jon P. Barker (Univ. of Sheffield)<br /> * Kenichi Kumatani (Spansion)<br /> * Jordan Cohen (Spelamode)<br /> * Leslie Smith (Univ. Stirling)<br /> * Punit Kumar Mohanty (Tata Consultancy Services)<br /> * Gérard Chollet (Télécom ParisTech)<br /> * Otman Chakkor (Univ. Tetuan)<br /> * Ngoc Q.K. Duong (Technicolor)<br /> * George Sterpu (Triniy College Dublin)<br /> * Zhijian Ou (Tsinghua Univ.)<br /> * Antti Hurmalainen (TU Tampere)<br /> * Tuomas Virtanen (TU Tampere)<br /> * Vinay Modi (Voice Pitara Technologies)<br /> * Jidong Tao (Vlingo)<br /> * Raphaël Blouet<br /> * Estelle Delpech</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=List_of_SIG_members&diff=460 List of SIG members 2016-10-13T08:32:12Z <p>Ramon.astudillo: </p> <hr /> <div>130 members as of September 9, 2016:<br /> * Søren Holdt Jensen (Univ. Aalborg)<br /> * Børge Lindberg (Univ. Aalborg)<br /> * Zhen-Hua Tan (Univ. Aalborg)<br /> * Kalle Palomäki (Aalto Univ.)<br /> * Steve Beet (Aculab)<br /> * Gautham Mysore (Adobe)<br /> * Björn Hoffmeister (Amazon)<br /> * Dimitris Dimitriadis (AT&amp;T Research)<br /> * Welly Naptali (ATR-Trek <br /> * Nassos Katsamanis (TU Athens)<br /> * David Pearce (Audience)<br /> * Keith Ponting (Avaya)<br /> * Daniele Giacobello (Beats by Dr. Dre)<br /> * Armin Sehr (Beuth HS)<br /> * Morgan Nelson (Univ. California, Berkeley)<br /> * Lalan Kumar (IIT Bhubaneswar)<br /> * Pranav Jawale (IIT Bombay)<br /> * Rajbabu Velmurugan (IIT Bombay)<br /> * Kateřina Žmolíková (Brno University of Technology)<br /> * Mark Gales (Univ. Cambrigde)<br /> * Angel Mario Castro Martínez (Univ. Carl von Ossietzky)<br /> * Bhiksha Raj (Carnegie-Mellon Univ.)<br /> * Rita Singh (Carnegie-Mellon Univ.)<br /> * Richard Stern (Carnegie-Mellon Univ.)<br /> * Climent Nadeu (UP Catalonia)<br /> * Dayana Ribas Gonzalez (CENATAV)<br /> * Nestor Becerra Yoma (Univ. Chile)<br /> * Wei Xue (Chinese Academy of Sciences)<br /> * Orpheus Wang (Chinese Academy of Sciences)<br /> * Fabio Tesser (CNR)<br /> * Jahangir Alam (CRIM)<br /> * Gerhard Taljaard (CSIR)<br /> * Petr Mizera (Czech TU)<br /> * Harish-Chandra Dubey (UT Dallas)<br /> * John L. Hansen (UT Dallas)<br /> * Taufiq Hasan (UT Dallas)<br /> * Avni Rajpal (IICT Dhirubhai Ambani)<br /> * Alexey Tarasov (Dublin IT)<br /> * Emanuël Habets (Univ. Erlangen-Nürnberg)<br /> * Walter Kellermann (Univ. Erlangen-Nürnberg)<br /> * Roland Maas (Univ. Erlangen-Nürnberg)<br /> * Georg Stemmer (Univ. Erlangen-Nürnberg)<br /> * Alessio Brutti (FBK)<br /> * Cristina Guerrero Flores (FBK)<br /> * Waheed Mohamed (FBK)<br /> * Brigitte Richardson (Ford Motors)<br /> * Thad Hughes (Google)<br /> * Ananya Misra (Google)<br /> * Arun Narayanan (Google)<br /> * Rohit Prabhavalkar (Google)<br /> * Kevin Wilson (Google)<br /> * José Andrés Gonzalez (Univ. Granada)<br /> * Juan Andres Morales Cordovilla (Univ. Granada)<br /> * Antonio M. Peinado (Univ. Granada)<br /> * Pejman Mowlaee (Univ. Graz)<br /> * Kwang Myung Jeon (Gwangju IST)<br /> * Martin Heckmann (Honda)<br /> * Afsaneh Asaei (Idiap)<br /> * Ramya Rasipuram (Idiap)<br /> * Alberto Abad (INESC-Id)<br /> * Ramon F. Astudillo (INESC-Id)<br /> * Sunit Sivasankaran (Inria)<br /> * Emmanuel Vincent (Inria)<br /> * Nick Ruiz (Interactions, LLC)<br /> * João Felipe Santos (INRS-EMT)<br /> * Rohit Gowtham (Interactive Intelligence)<br /> * Hynek Hermansky (Johns-Hopkins Univ.)<br /> * Vijayaditya Peddinti (Johns-Hopkins Univ.)<br /> * Rajesh Hegde (IIT Kanpur)<br /> * Karan Nathwani (IIT Kanpur)<br /> * Jort Gemmeke (KU Leuven)<br /> * Hugo van Hamme (KU Leuven)<br /> * Jiri Malek (Univ. Liberec)<br /> * Cong-Thanh Do (LIMSI)<br /> * Arseniy Gorin (LIMSI)<br /> * Rajiv Reddy (LumenVox)<br /> * Alfred Mertins (Univ. Lübeck)<br /> * Florian Müller (Univ. Lübeck)<br /> * Ke Chen (Univ. Manchester)<br /> * Richard Rose (McGill Univ.)<br /> * John Hershey (MERL)<br /> * Jonathan Le Roux (MERL)<br /> * Shinji Watanabe (MERL)<br /> * Mike Seltzer (Microsoft)<br /> * Joseph P. Campbell (MIT)<br /> * Jim Glass (MIT)<br /> * Najim Dehak (MIT)<br /> * Björn Schuller (TU Munich)<br /> * Felix Weninger (TU Munich)<br /> * Xiong Xiao (Nanyang Technological Univ.)<br /> * Lou Boves (Univ. Nijmegen)<br /> * Mario Ganzeboom (Univ. Nijmegen)<br /> * Bert Cranen (Univ. Nijmegen)<br /> * Louis ten Bosch (Univ. Nijmegen)<br /> * David van Leeuwen (Univ. Nijmegen)<br /> * Niksa Jakovljevic (Univ. Novi Sad)<br /> * Shuguang Wang (Northwestern Polytechnical Univ., Xi'an)<br /> * Marc Delcroix (NTT)<br /> * Miquel Espi (NTT)<br /> * Keisuke Kinoshita (NTT)<br /> * Dung Tran (NTT)<br /> * Cyril Guillaumé (NXP Software)<br /> * Wouter Tirry (NXP Software)<br /> * DeLiang Wang (Ohio State Univ.)<br /> * Zhongqiu Wang (Ohio State Univ.)<br /> * Simon Bozonnet (OnMobile)<br /> * Anil Alexander (Oxford Wave Research)<br /> * Reinhold Häb-Umbach (Univ. Paderborn)<br /> * Ricard Marxer (Univ. of Sheffield)<br /> * Ahmed Hussen Abdelaziz (Ruhr Univ., Bochum)<br /> * Dorothea Kolossa (Ruhr Univ., Bochum)<br /> * Tobias Menne (RWTH University, Aachen)<br /> * Rahil Mahdiyan (Saarland Univ.)<br /> * Hakan Erdogan (Sabanci Univ.)<br /> * Emad Grais (Sabanci Univ.)<br /> * Raghavendra Bilgi (Samsung)<br /> * David Trainor (Sentireal)<br /> * Jon P. Barker (Univ. of Sheffield)<br /> * Kenichi Kumatani (Spansion)<br /> * Jordan Cohen (Spelamode)<br /> * Leslie Smith (Univ. Stirling)<br /> * Punit Kumar Mohanty (Tata Consultancy Services)<br /> * Gérard Chollet (Télécom ParisTech)<br /> * Otman Chakkor (Univ. Tetuan)<br /> * Ngoc Q.K. Duong (Technicolor)<br /> * Zhijian Ou (Tsinghua Univ.)<br /> * Antti Hurmalainen (TU Tampere)<br /> * Tuomas Virtanen (TU Tampere)<br /> * Vinay Modi (Voice Pitara Technologies)<br /> * Jidong Tao (Vlingo)<br /> * Raphaël Blouet<br /> * Estelle Delpech</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=List_of_SIG_members&diff=459 List of SIG members 2016-10-13T08:31:23Z <p>Ramon.astudillo: </p> <hr /> <div>130 members as of September 9, 2016:<br /> * Søren Holdt Jensen (Univ. Aalborg)<br /> * Børge Lindberg (Univ. Aalborg)<br /> * Zhen-Hua Tan (Univ. Aalborg)<br /> * Kalle Palomäki (Aalto Univ.)<br /> * Steve Beet (Aculab)<br /> * Gautham Mysore (Adobe)<br /> * Björn Hoffmeister (Amazon)<br /> * Dimitris Dimitriadis (AT&amp;T Research)<br /> * Welly Naptali (ATR-Trek <br /> * Nassos Katsamanis (TU Athens)<br /> * David Pearce (Audience)<br /> * Keith Ponting (Avaya)<br /> * Daniele Giacobello (Beats by Dr. Dre)<br /> * Armin Sehr (Beuth HS)<br /> * Morgan Nelson (Univ. California, Berkeley)<br /> * Lalan Kumar (IIT Bhubaneswar)<br /> * Pranav Jawale (IIT Bombay)<br /> * Rajbabu Velmurugan (IIT Bombay)<br /> * Kateřina Žmolíková (Brno University of Technology)<br /> * Mark Gales (Univ. Cambrigde)<br /> * Angel Mario Castro Martínez (Univ. Carl von Ossietzky)<br /> * Bhiksha Raj (Carnegie-Mellon Univ.)<br /> * Rita Singh (Carnegie-Mellon Univ.)<br /> * Richard Stern (Carnegie-Mellon Univ.)<br /> * Climent Nadeu (UP Catalonia)<br /> * Dayana Ribas Gonzalez (CENATAV)<br /> * Avni Rajpal (IICT Dhirubhai Ambani)<br /> * Nestor Becerra Yoma (Univ. Chile)<br /> * Wei Xue (Chinese Academy of Sciences)<br /> * Orpheus Wang (Chinese Academy of Sciences)<br /> * Fabio Tesser (CNR)<br /> * Jahangir Alam (CRIM)<br /> * Gerhard Taljaard (CSIR)<br /> * Petr Mizera (Czech TU)<br /> * Harish-Chandra Dubey (UT Dallas)<br /> * John L. Hansen (UT Dallas)<br /> * Taufiq Hasan (UT Dallas)<br /> * Alexey Tarasov (Dublin IT)<br /> * Emanuël Habets (Univ. Erlangen-Nürnberg)<br /> * Walter Kellermann (Univ. Erlangen-Nürnberg)<br /> * Roland Maas (Univ. Erlangen-Nürnberg)<br /> * Georg Stemmer (Univ. Erlangen-Nürnberg)<br /> * Alessio Brutti (FBK)<br /> * Cristina Guerrero Flores (FBK)<br /> * Waheed Mohamed (FBK)<br /> * Brigitte Richardson (Ford Motors)<br /> * Thad Hughes (Google)<br /> * Ananya Misra (Google)<br /> * Arun Narayanan (Google)<br /> * Rohit Prabhavalkar (Google)<br /> * Kevin Wilson (Google)<br /> * José Andrés Gonzalez (Univ. Granada)<br /> * Juan Andres Morales Cordovilla (Univ. Granada)<br /> * Antonio M. Peinado (Univ. Granada)<br /> * Pejman Mowlaee (Univ. Graz)<br /> * Kwang Myung Jeon (Gwangju IST)<br /> * Martin Heckmann (Honda)<br /> * Afsaneh Asaei (Idiap)<br /> * Ramya Rasipuram (Idiap)<br /> * Alberto Abad (INESC-Id)<br /> * Ramon F. Astudillo (INESC-Id)<br /> * Sunit Sivasankaran (Inria)<br /> * Emmanuel Vincent (Inria)<br /> * Nick Ruiz (Interactions, LLC)<br /> * João Felipe Santos (INRS-EMT)<br /> * Rohit Gowtham (Interactive Intelligence)<br /> * Hynek Hermansky (Johns-Hopkins Univ.)<br /> * Vijayaditya Peddinti (Johns-Hopkins Univ.)<br /> * Rajesh Hegde (IIT Kanpur)<br /> * Karan Nathwani (IIT Kanpur)<br /> * Jort Gemmeke (KU Leuven)<br /> * Hugo van Hamme (KU Leuven)<br /> * Jiri Malek (Univ. Liberec)<br /> * Cong-Thanh Do (LIMSI)<br /> * Arseniy Gorin (LIMSI)<br /> * Rajiv Reddy (LumenVox)<br /> * Alfred Mertins (Univ. Lübeck)<br /> * Florian Müller (Univ. Lübeck)<br /> * Ke Chen (Univ. Manchester)<br /> * Richard Rose (McGill Univ.)<br /> * John Hershey (MERL)<br /> * Jonathan Le Roux (MERL)<br /> * Shinji Watanabe (MERL)<br /> * Mike Seltzer (Microsoft)<br /> * Joseph P. Campbell (MIT)<br /> * Jim Glass (MIT)<br /> * Najim Dehak (MIT)<br /> * Björn Schuller (TU Munich)<br /> * Felix Weninger (TU Munich)<br /> * Xiong Xiao (Nanyang Technological Univ.)<br /> * Lou Boves (Univ. Nijmegen)<br /> * Mario Ganzeboom (Univ. Nijmegen)<br /> * Bert Cranen (Univ. Nijmegen)<br /> * Louis ten Bosch (Univ. Nijmegen)<br /> * David van Leeuwen (Univ. Nijmegen)<br /> * Niksa Jakovljevic (Univ. Novi Sad)<br /> * Shuguang Wang (Northwestern Polytechnical Univ., Xi'an)<br /> * Marc Delcroix (NTT)<br /> * Miquel Espi (NTT)<br /> * Keisuke Kinoshita (NTT)<br /> * Dung Tran (NTT)<br /> * Cyril Guillaumé (NXP Software)<br /> * Wouter Tirry (NXP Software)<br /> * DeLiang Wang (Ohio State Univ.)<br /> * Zhongqiu Wang (Ohio State Univ.)<br /> * Simon Bozonnet (OnMobile)<br /> * Anil Alexander (Oxford Wave Research)<br /> * Reinhold Häb-Umbach (Univ. Paderborn)<br /> * Ricard Marxer (Univ. of Sheffield)<br /> * Ahmed Hussen Abdelaziz (Ruhr Univ., Bochum)<br /> * Dorothea Kolossa (Ruhr Univ., Bochum)<br /> * Tobias Menne (RWTH University, Aachen)<br /> * Rahil Mahdiyan (Saarland Univ.)<br /> * Hakan Erdogan (Sabanci Univ.)<br /> * Emad Grais (Sabanci Univ.)<br /> * Raghavendra Bilgi (Samsung)<br /> * David Trainor (Sentireal)<br /> * Jon P. Barker (Univ. of Sheffield)<br /> * Kenichi Kumatani (Spansion)<br /> * Jordan Cohen (Spelamode)<br /> * Leslie Smith (Univ. Stirling)<br /> * Punit Kumar Mohanty (Tata Consultancy Services)<br /> * Gérard Chollet (Télécom ParisTech)<br /> * Otman Chakkor (Univ. Tetuan)<br /> * Ngoc Q.K. Duong (Technicolor)<br /> * Zhijian Ou (Tsinghua Univ.)<br /> * Antti Hurmalainen (TU Tampere)<br /> * Tuomas Virtanen (TU Tampere)<br /> * Vinay Modi (Voice Pitara Technologies)<br /> * Jidong Tao (Vlingo)<br /> * Raphaël Blouet<br /> * Estelle Delpech</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=List_of_SIG_members&diff=448 List of SIG members 2016-06-20T21:38:23Z <p>Ramon.astudillo: </p> <hr /> <div>128 members as of May 2, 2016:<br /> * Søren Holdt Jensen (Univ. Aalborg)<br /> * Børge Lindberg (Univ. Aalborg)<br /> * Zhen-Hua Tan (Univ. Aalborg)<br /> * Kalle Palomäki (Aalto Univ.)<br /> * Steve Beet (Aculab)<br /> * Gautham Mysore (Adobe)<br /> * Björn Hoffmeister (Amazon)<br /> * Dimitris Dimitriadis (AT&amp;T Research)<br /> * Welly Naptali (ATR-Trek Co.)<br /> * Nassos Katsamanis (TU Athens)<br /> * David Pearce (Audience)<br /> * Keith Ponting (Avaya)<br /> * Daniele Giacobello (Beats by Dr. Dre)<br /> * Armin Sehr (Beuth HS)<br /> * Morgan Nelson (Univ. California, Berkeley)<br /> * Pranav Jawale (IIT Bombay)<br /> * Rajbabu Velmurugan (IIT Bombay)<br /> * Kateřina Žmolíková (Brno University of Technology)<br /> * Mark Gales (Univ. Cambrigde)<br /> * Angel Mario Castro Martínez (Univ. Carl von Ossietzky)<br /> * Bhiksha Raj (Carnegie-Mellon Univ.)<br /> * Rita Singh (Carnegie-Mellon Univ.)<br /> * Richard Stern (Carnegie-Mellon Univ.)<br /> * Climent Nadeu (UP Catalonia)<br /> * Dayana Ribas Gonzalez (CENATAV)<br /> * Nestor Becerra Yoma (Univ. Chile)<br /> * Wei Xue (Chinese Academy of Sciences)<br /> * Orpheus Wang (Chinese Academy of Sciences)<br /> * Fabio Tesser (CNR)<br /> * Jahangir Alam (CRIM)<br /> * Gerhard Taljaard (CSIR)<br /> * Petr Mizera (Czech TU)<br /> * Harish-Chandra Dubey (UT Dallas)<br /> * John L. Hansen (UT Dallas)<br /> * Taufiq Hasan (UT Dallas)<br /> * Alexey Tarasov (Dublin IT)<br /> * Emanuël Habets (Univ. Erlangen-Nürnberg)<br /> * Walter Kellermann (Univ. Erlangen-Nürnberg)<br /> * Roland Maas (Univ. Erlangen-Nürnberg)<br /> * Georg Stemmer (Univ. Erlangen-Nürnberg)<br /> * Alessio Brutti (FBK)<br /> * Cristina Guerrero Flores (FBK)<br /> * Waheed Mohamed (FBK)<br /> * Brigitte Richardson (Ford Motors)<br /> * Thad Hughes (Google)<br /> * Kevin Wilson (Google)<br /> * Rohit Prabhavalkar (Google)<br /> * José Andrés Gonzalez (Univ. Granada)<br /> * Antonio M. Peinado (Univ. Granada)<br /> * Pejman Mowlaee (Univ. Graz)<br /> * Kwang Myung Jeon (Gwangju IST)<br /> * Martin Heckmann (Honda)<br /> * Afsaneh Asaei (Idiap)<br /> * Ramya Rasipuram (Idiap)<br /> * Alberto Abad (INESC-Id)<br /> * Ramon F. Astudillo (INESC-Id)<br /> * Arseniy Gorin (Inria)<br /> * Juan Andres Morales Cordovilla (Inria)<br /> * Yann Salaün (Inria)<br /> * Sunit Sivasankaran (Inria)<br /> * Dung Tran (Inria)<br /> * Emmanuel Vincent (Inria)<br /> * Nick Ruiz (Interactions, LLC)<br /> * João Felipe Santos (INRS-EMT)<br /> * Rohit Gowtham (Interactive Intelligence)<br /> * Hynek Hermansky (Johns-Hopkins Univ.)<br /> * Vijayaditya Peddinti (Johns-Hopkins Univ.)<br /> * Rajesh Hegde (IIT Kanpur)<br /> * Lalan Kumar (IIT Kanpur)<br /> * Karan Nathwani (IIT Kanpur)<br /> * Jort Gemmeke (KU Leuven)<br /> * Hugo van Hamme (KU Leuven)<br /> * Jiri Malek (Univ. Liberec)<br /> * Cong-Thanh Do (LIMSI)<br /> * Rajiv Reddy (LumenVox)<br /> * Alfred Mertins (Univ. Lübeck)<br /> * Florian Müller (Univ. Lübeck)<br /> * Ke Chen (Univ. Manchester)<br /> * Richard Rose (McGill Univ.)<br /> * John Hershey (MERL)<br /> * Jonathan Le Roux (MERL)<br /> * Shinji Watanabe (MERL)<br /> * Mike Seltzer (Microsoft)<br /> * Joseph P. Campbell (MIT)<br /> * Jim Glass (MIT)<br /> * Najim Dehak (MIT)<br /> * Björn Schuller (TU Munich)<br /> * Felix Weninger (TU Munich)<br /> * Lou Boves (Univ. Nijmegen)<br /> * Mario Ganzeboom (Univ. Nijmegen)<br /> * Bert Cranen (Univ. Nijmegen)<br /> * Louis ten Bosch (Univ. Nijmegen)<br /> * David van Leeuwen (Univ. Nijmegen)<br /> * Niksa Jakovljevic (Univ. Novi Sad)<br /> * Shuguang Wang (Northwestern Polytechnical Univ., Xi'an)<br /> * Marc Delcroix (NTT)<br /> * Miquel Espi (NTT)<br /> * Keisuke Kinoshita (NTT)<br /> * Cyril Guillaumé (NXP Software)<br /> * Wouter Tirry (NXP Software)<br /> * Arun Narayanan (Ohio State Univ.)<br /> * DeLiang Wang (Ohio State Univ.)<br /> * Zhongqiu Wang (Ohio State Univ.)<br /> * Simon Bozonnet (OnMobile)<br /> * Anil Alexander (Oxford Wave Research)<br /> * Reinhold Häb-Umbach (Univ. Paderborn)<br /> * Ananya Misra (Princeton Univ.)<br /> * Ricard Marxer (Univ. of Sheffield)<br /> * Ahmed Hussen Abdelaziz (Ruhr Univ., Bochum)<br /> * Dorothea Kolossa (Ruhr Univ., Bochum)<br /> * Tobias Menne (RWTH University, Aachen)<br /> * Rahil Mahdiyan (Saarland Univ.)<br /> * Hakan Erdogan (Sabanci Univ.)<br /> * Emad Grais (Sabanci Univ.)<br /> * Raghavendra Bilgi (Samsung)<br /> * David Trainor (Sentireal)<br /> * Jon P. Barker (Univ. of Sheffield)<br /> * Kenichi Kumatani (Spansion)<br /> * Jordan Cohen (Spelamode)<br /> * Leslie Smith (Univ. Stirling)<br /> * Punit Kumar Mohanty (Tata Consultancy Services)<br /> * Gérard Chollet (Télécom ParisTech)<br /> * Otman Chakkor (Univ. Tetuan)<br /> * Ngoc Q.K. Duong (Technicolor)<br /> * Zhijian Ou (Tsinghua Univ.)<br /> * Antti Hurmalainen (TU Tampere)<br /> * Tuomas Virtanen (TU Tampere)<br /> * Vinay Modi (Voice Pitara Technologies)<br /> * Jidong Tao (Vlingo)<br /> * Raphaël Blouet</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=List_of_SIG_members&diff=441 List of SIG members 2016-04-26T19:01:25Z <p>Ramon.astudillo: </p> <hr /> <div>124 members as of March 22, 2016:<br /> * Søren Holdt Jensen (Univ. Aalborg)<br /> * Børge Lindberg (Univ. Aalborg)<br /> * Zhen-Hua Tan (Univ. Aalborg)<br /> * Kalle Palomäki (Aalto Univ.)<br /> * Steve Beet (Aculab)<br /> * Gautham Mysore (Adobe)<br /> * Björn Hoffmeister (Amazon)<br /> * Dimitris Dimitriadis (AT&amp;T Research)<br /> * Welly Naptali (ATR-Trek Co.)<br /> * Nassos Katsamanis (TU Athens)<br /> * David Pearce (Audience)<br /> * Wei Xue (Institution of Automation, Chinese Academy of Sciences)<br /> * Keith Ponting (Avaya)<br /> * Daniele Giacobello (Beats by Dr. Dre)<br /> * Armin Sehr (Beuth HS)<br /> * Morgan Nelson (Univ. California, Berkeley)<br /> * Pranav Jawale (IIT Bombay)<br /> * Rajbabu Velmurugan (IIT Bombay)<br /> * Kateřina Žmolíková (Brno University of Technology)<br /> * Mark Gales (Univ. Cambrigde)<br /> * Angel Mario Castro Martínez (Univ. Carl von Ossietzky)<br /> * Bhiksha Raj (Carnegie-Mellon Univ.)<br /> * Rita Singh (Carnegie-Mellon Univ.)<br /> * Richard Stern (Carnegie-Mellon Univ.)<br /> * Climent Nadeu (UP Catalonia)<br /> * Dayana Ribas Gonzalez (CENATAV)<br /> * Nestor Becerra Yoma (Univ. Chile)<br /> * Fabio Tesser (CNR)<br /> * Jahangir Alam (CRIM)<br /> * Gerhard Taljaard (CSIR)<br /> * Petr Mizera (Czech TU)<br /> * Harish-Chandra Dubey (UT Dallas)<br /> * John L. Hansen (UT Dallas)<br /> * Taufiq Hasan (UT Dallas)<br /> * Alexey Tarasov (Dublin IT)<br /> * Emanuël Habets (Univ. Erlangen-Nürnberg)<br /> * Walter Kellermann (Univ. Erlangen-Nürnberg)<br /> * Roland Maas (Univ. Erlangen-Nürnberg)<br /> * Georg Stemmer (Univ. Erlangen-Nürnberg)<br /> * Alessio Brutti (FBK)<br /> * Cristina Guerrero Flores (FBK)<br /> * Waheed Mohamed (FBK)<br /> * Brigitte Richardson (Ford Motors)<br /> * Thad Hughes (Google)<br /> * Kevin Wilson (Google)<br /> * Rohit Prabhavalkar (Google)<br /> * José Andrés Gonzalez (Univ. Granada)<br /> * Antonio M. Peinado (Univ. Granada)<br /> * Pejman Mowlaee (Univ. Graz)<br /> * Kwang Myung Jeon (Gwangju IST)<br /> * Martin Heckmann (Honda)<br /> * Afsaneh Asaei (Idiap)<br /> * Ramya Rasipuram (Idiap)<br /> * Alberto Abad (INESC-Id)<br /> * Ramon F. Astudillo (INESC-Id)<br /> * Arseniy Gorin (Inria)<br /> * Juan Andres Morales Cordovilla (Inria)<br /> * Yann Salaün (Inria)<br /> * Sunit Sivasankaran (Inria)<br /> * Dung Tran (Inria)<br /> * Emmanuel Vincent (Inria)<br /> * João Felipe Santos (INRS-EMT)<br /> * Rohit Gowtham (Interactive Intelligence)<br /> * Hynek Hermansky (Johns-Hopkins Univ.)<br /> * Peddinti Vijayaditya (Johns-Hopkins Univ.)<br /> * Rajesh Hegde (IIT Kanpur)<br /> * Lalan Kumar (IIT Kanpur)<br /> * Karan Nathwani (IIT Kanpur)<br /> * Jort Gemmeke (KU Leuven)<br /> * Hugo van Hamme (KU Leuven)<br /> * Jiri Malek (Univ. Liberec)<br /> * Cong-Thanh Do (LIMSI)<br /> * Rajiv Reddy (LumenVox)<br /> * Alfred Mertins (Univ. Lübeck)<br /> * Florian Müller (Univ. Lübeck)<br /> * Ke Chen (Univ. Manchester)<br /> * Richard Rose (McGill Univ.)<br /> * John Hershey (MERL)<br /> * Jonathan Le Roux (MERL)<br /> * Shinji Watanabe (MERL)<br /> * Mike Seltzer (Microsoft)<br /> * Joseph P. Campbell (MIT)<br /> * Jim Glass (MIT)<br /> * Najim Dehak (MIT)<br /> * Björn Schuller (TU Munich)<br /> * Felix Weninger (TU Munich)<br /> * Lou Boves (Univ. Nijmegen)<br /> * Mario Ganzeboom (Univ. Nijmegen)<br /> * Bert Cranen (Univ. Nijmegen)<br /> * Louis ten Bosch (Univ. Nijmegen)<br /> * David van Leeuwen (Univ. Nijmegen)<br /> * Niksa Jakovljevic (Univ. Novi Sad)<br /> * Shuguang Wang (Northwestern Polytechnical Univ., Xi'an)<br /> * Marc Delcroix (NTT)<br /> * Miquel Espi (NTT)<br /> * Keisuke Kinoshita (NTT)<br /> * Cyril Guillaumé (NXP Software)<br /> * Wouter Tirry (NXP Software)<br /> * Arun Narayanan (Ohio State Univ.)<br /> * DeLiang Wang (Ohio State Univ.)<br /> * Zhongqiu Wang (Ohio State Univ.)<br /> * Simon Bozonnet (OnMobile)<br /> * Anil Alexander (Oxford Wave Research)<br /> * Reinhold Häb-Umbach (Univ. Paderborn)<br /> * Ananya Misra (Princeton Univ.)<br /> * Ricard Marxer (Univ. of Sheffield)<br /> * Ahmed Hussen Abdelaziz (Ruhr Univ., Bochum)<br /> * Dorothea Kolossa (Ruhr Univ., Bochum)<br /> * Tobias Menne (RWTH University, Aachen)<br /> * Rahil Mahdiyan (Saarland Univ.)<br /> * Hakan Erdogan (Sabanci Univ.)<br /> * Emad Grais (Sabanci Univ.)<br /> * Raghavendra Bilgi (Samsung)<br /> * David Trainor (Sentireal)<br /> * Kenichi Kumatani (Spansion)<br /> * Jordan Cohen (Spelamode)<br /> * Leslie Smith (Univ. Stirling)<br /> * Gérard Chollet (Télécom ParisTech)<br /> * Otman Chakkor (Univ. Tetuan)<br /> * Ngoc Q.K. Duong (Technicolor)<br /> * Zhijian Ou (Tsinghua Univ.)<br /> * Antti Hurmalainen (TU Tampere)<br /> * Tuomas Virtanen (TU Tampere)<br /> * Vinay Modi (Voice Pitara Technologies)<br /> * Jidong Tao (Vlingo)<br /> * Raphaël Blouet</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=List_of_SIG_members&diff=422 List of SIG members 2015-10-14T10:13:54Z <p>Ramon.astudillo: </p> <hr /> <div>120 members as of October 13, 2015:<br /> * Søren Holdt Jensen (Univ. Aalborg)<br /> * Børge Lindberg (Univ. Aalborg)<br /> * Zhen-Hua Tan (Univ. Aalborg)<br /> * Kalle Palomäki (Aalto Univ.)<br /> * Steve Beet (Aculab)<br /> * Gautham Mysore (Adobe)<br /> * Björn Hoffmeister (Amazon)<br /> * Dimitris Dimitriadis (AT&amp;T Research)<br /> * Welly Naptali (ATR-Trek Co.)<br /> * Nassos Katsamanis (TU Athens)<br /> * David Pearce (Audience)<br /> * Wei Xue (Institution of Automation, Chinese Academy of Sciences)<br /> * Keith Ponting (Avaya)<br /> * Daniele Giacobello (Beats by Dr. Dre)<br /> * Armin Sehr (Beuth HS)<br /> * Morgan Nelson (Univ. California, Berkeley)<br /> * Pranav Jawale (IIT Bombay)<br /> * Rajbabu Velmurugan (IIT Bombay)<br /> * Kateřina Žmolíková (Brno University of Technology)<br /> * Mark Gales (Univ. Cambrigde)<br /> * Angel Mario Castro Martínez (Univ. Carl von Ossietzky)<br /> * Bhiksha Raj (Carnegie-Mellon Univ.)<br /> * Rita Singh (Carnegie-Mellon Univ.)<br /> * Richard Stern (Carnegie-Mellon Univ.)<br /> * Climent Nadeu (UP Catalonia)<br /> * Nestor Becerra Yoma (Univ. Chile)<br /> * Fabio Tesser (CNR)<br /> * Jahangir Alam (CRIM)<br /> * Gerhard Taljaard (CSIR)<br /> * John L. Hansen (UT Dallas)<br /> * Taufiq Hasan (UT Dallas)<br /> * Alexey Tarasov (Dublin IT)<br /> * Emanuël Habets (Univ. Erlangen-Nürnberg)<br /> * Walter Kellermann (Univ. Erlangen-Nürnberg)<br /> * Roland Maas (Univ. Erlangen-Nürnberg)<br /> * Georg Stemmer (Univ. Erlangen-Nürnberg)<br /> * Alessio Brutti (FBK)<br /> * Cristina Guerrero Flores (FBK)<br /> * Waheed Mohamed (FBK)<br /> * Brigitte Richardson (Ford Motors)<br /> * Thad Hughes (Google)<br /> * Kevin Wilson (Google)<br /> * Rohit Prabhavalkar (Google)<br /> * José Andrés Gonzalez (Univ. Granada)<br /> * Antonio M. Peinado (Univ. Granada)<br /> * Pejman Mowlaee (Univ. Graz)<br /> * Kwang Myung Jeon (Gwangju IST)<br /> * Martin Heckmann (Honda)<br /> * Afsaneh Asaei (Idiap)<br /> * Ramya Rasipuram (Idiap)<br /> * Alberto Abad (INESC-Id)<br /> * Ramon F. Astudillo (INESC-Id)<br /> * Arseniy Gorin (Inria)<br /> * Juan Andres Morales Cordovilla (Inria)<br /> * Yann Salaün (Inria)<br /> * Sunit Sivasankaran (Inria)<br /> * Dung Tran (Inria)<br /> * Emmanuel Vincent (Inria)<br /> * João Felipe Santos (INRS-EMT)<br /> * Rohit Gowtham (Interactive Intelligence)<br /> * Hynek Hermansky (Johns-Hopkins Univ.)<br /> * Rajesh Hegde (IIT Kanpur)<br /> * Lalan Kumar (IIT Kanpur)<br /> * Karan Nathwani (IIT Kanpur)<br /> * Jort Gemmeke (KU Leuven)<br /> * Hugo van Hamme (KU Leuven)<br /> * Jiri Malek (Univ. Liberec)<br /> * Cong-Thanh Do (LIMSI)<br /> * Rajiv Reddy (LumenVox)<br /> * Alfred Mertins (Univ. Lübeck)<br /> * Florian Müller (Univ. Lübeck)<br /> * Ke Chen (Univ. Manchester)<br /> * Richard Rose (McGill Univ.)<br /> * John Hershey (MERL)<br /> * Jonathan Le Roux (MERL)<br /> * Shinji Watanabe (MERL)<br /> * Mike Seltzer (Microsoft)<br /> * Joseph P. Campbell (MIT)<br /> * Jim Glass (MIT)<br /> * Najim Dehak (MIT)<br /> * Björn Schuller (TU Munich)<br /> * Felix Weninger (TU Munich)<br /> * Lou Boves (Univ. Nijmegen)<br /> * Mario Ganzeboom (Univ. Nijmegen)<br /> * Bert Cranen (Univ. Nijmegen)<br /> * Louis ten Bosch (Univ. Nijmegen)<br /> * David van Leeuwen (Univ. Nijmegen)<br /> * Niksa Jakovljevic (Univ. Novi Sad)<br /> * Shuguang Wang (Northwestern Polytechnical Univ., Xi'an)<br /> * Marc Delcroix (NTT)<br /> * Miquel Espi (NTT)<br /> * Keisuke Kinoshita (NTT)<br /> * Cyril Guillaumé (NXP Software)<br /> * Wouter Tirry (NXP Software)<br /> * Arun Narayanan (Ohio State Univ.)<br /> * DeLiang Wang (Ohio State Univ.)<br /> * Simon Bozonnet (OnMobile)<br /> * Anil Alexander (Oxford Wave Research)<br /> * Reinhold Häb-Umbach (Univ. Paderborn)<br /> * Ananya Misra (Princeton Univ.)<br /> * Ricard Marxer (Univ. of Sheffield)<br /> * Ahmed Hussen Abdelaziz (Ruhr Univ., Bochum)<br /> * Dorothea Kolossa (Ruhr Univ., Bochum)<br /> * Tobias Menne (RWTH University, Aachen)<br /> * Rahil Mahdiyan (Saarland Univ.)<br /> * Hakan Erdogan (Sabanci Univ.)<br /> * Emad Grais (Sabanci Univ.)<br /> * Raghavendra Bilgi (Samsung)<br /> * David Trainor (Sentireal)<br /> * Kenichi Kumatani (Spansion)<br /> * Jordan Cohen (Spelamode)<br /> * Leslie Smith (Univ. Stirling)<br /> * Gérard Chollet (Télécom ParisTech)<br /> * Otman Chakkor (Univ. Tetuan)<br /> * Ngoc Q.K. Duong (Technicolor)<br /> * Zhijian Ou (Tsinghua Univ.)<br /> * Antti Hurmalainen (TU Tampere)<br /> * Tuomas Virtanen (TU Tampere)<br /> * Vinay Modi (Voice Pitara Technologies)<br /> * Jidong Tao (Vlingo)<br /> * Raphaël Blouet</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=List_of_SIG_members&diff=420 List of SIG members 2015-10-12T20:50:52Z <p>Ramon.astudillo: </p> <hr /> <div>119 members as of October 10, 2015:<br /> * Søren Holdt Jensen (Univ. Aalborg)<br /> * Børge Lindberg (Univ. Aalborg)<br /> * Zhen-Hua Tan (Univ. Aalborg)<br /> * Kalle Palomäki (Aalto Univ.)<br /> * Steve Beet (Aculab)<br /> * Gautham Mysore (Adobe)<br /> * Björn Hoffmeister (Amazon)<br /> * Dimitris Dimitriadis (AT&amp;T Research)<br /> * Welly Naptali (ATR-Trek Co.)<br /> * Nassos Katsamanis (TU Athens)<br /> * David Pearce (Audience)<br /> * Wei Xue (Institution of Automation, Chinese Academy of Sciences)<br /> * Keith Ponting (Avaya)<br /> * Daniele Giacobello (Beats by Dr. Dre)<br /> * Armin Sehr (Beuth HS)<br /> * Morgan Nelson (Univ. California, Berkeley)<br /> * Pranav Jawale (IIT Bombay)<br /> * Rajbabu Velmurugan (IIT Bombay)<br /> * Kateřina Žmolíková (Brno University of Technology)<br /> * Mark Gales (Univ. Cambrigde)<br /> * Angel Mario Castro Martínez (Univ. Carl von Ossietzky)<br /> * Bhiksha Raj (Carnegie-Mellon Univ.)<br /> * Rita Singh (Carnegie-Mellon Univ.)<br /> * Richard Stern (Carnegie-Mellon Univ.)<br /> * Climent Nadeu (UP Catalonia)<br /> * Nestor Becerra Yoma (Univ. Chile)<br /> * Fabio Tesser (CNR)<br /> * Jahangir Alam (CRIM)<br /> * Gerhard Taljaard (CSIR)<br /> * John L. Hansen (UT Dallas)<br /> * Taufiq Hasan (UT Dallas)<br /> * Alexey Tarasov (Dublin IT)<br /> * Emanuël Habets (Univ. Erlangen-Nürnberg)<br /> * Walter Kellermann (Univ. Erlangen-Nürnberg)<br /> * Roland Maas (Univ. Erlangen-Nürnberg)<br /> * Georg Stemmer (Univ. Erlangen-Nürnberg)<br /> * Alessio Brutti (FBK)<br /> * Cristina Guerrero Flores (FBK)<br /> * Waheed Mohamed (FBK)<br /> * Brigitte Richardson (Ford Motors)<br /> * Thad Hughes (Google)<br /> * Kevin Wilson (Google)<br /> * Rohit Prabhavalkar (Google)<br /> * José Andrés Gonzalez (Univ. Granada)<br /> * Antonio M. Peinado (Univ. Granada)<br /> * Pejman Mowlaee (Univ. Graz)<br /> * Kwang Myung Jeon (Gwangju IST)<br /> * Martin Heckmann (Honda)<br /> * Afsaneh Asaei (Idiap)<br /> * Ramya Rasipuram (Idiap)<br /> * Alberto Abad (INESC-Id)<br /> * Ramon F. Astudillo (INESC-Id)<br /> * Arseniy Gorin (Inria)<br /> * Juan Andres Morales Cordovilla (Inria)<br /> * Yann Salaün (Inria)<br /> * Sunit Sivasankaran (Inria)<br /> * Dung Tran (Inria)<br /> * Emmanuel Vincent (Inria)<br /> * João Felipe Santos (INRS-EMT)<br /> * Rohit Gowtham (Interactive Intelligence)<br /> * Hynek Hermansky (Johns-Hopkins Univ.)<br /> * Rajesh Hegde (IIT Kanpur)<br /> * Lalan Kumar (IIT Kanpur)<br /> * Karan Nathwani (IIT Kanpur)<br /> * Jort Gemmeke (KU Leuven)<br /> * Hugo van Hamme (KU Leuven)<br /> * Jiri Malek (Univ. Liberec)<br /> * Cong-Thanh Do (LIMSI)<br /> * Rajiv Reddy (LumenVox)<br /> * Alfred Mertins (Univ. Lübeck)<br /> * Florian Müller (Univ. Lübeck)<br /> * Ke Chen (Univ. Manchester)<br /> * Richard Rose (McGill Univ.)<br /> * John Hershey (MERL)<br /> * Jonathan Le Roux (MERL)<br /> * Shinji Watanabe (MERL)<br /> * Mike Seltzer (Microsoft)<br /> * Joseph P. Campbell (MIT)<br /> * Jim Glass (MIT)<br /> * Najim Dehak (MIT)<br /> * Björn Schuller (TU Munich)<br /> * Felix Weninger (TU Munich)<br /> * Lou Boves (Univ. Nijmegen)<br /> * Mario Ganzeboom (Univ. Nijmegen)<br /> * Bert Cranen (Univ. Nijmegen)<br /> * Louis ten Bosch (Univ. Nijmegen)<br /> * David van Leeuwen (Univ. Nijmegen)<br /> * Shuguang Wang (Northwestern Polytechnical Univ., Xi'an)<br /> * Marc Delcroix (NTT)<br /> * Miquel Espi (NTT)<br /> * Keisuke Kinoshita (NTT)<br /> * Cyril Guillaumé (NXP Software)<br /> * Wouter Tirry (NXP Software)<br /> * Arun Narayanan (Ohio State Univ.)<br /> * DeLiang Wang (Ohio State Univ.)<br /> * Simon Bozonnet (OnMobile)<br /> * Anil Alexander (Oxford Wave Research)<br /> * Reinhold Häb-Umbach (Univ. Paderborn)<br /> * Ananya Misra (Princeton Univ.)<br /> * Ricard Marxer (Univ. of Sheffield)<br /> * Ahmed Hussen Abdelaziz (Ruhr Univ., Bochum)<br /> * Dorothea Kolossa (Ruhr Univ., Bochum)<br /> * Tobias Menne (RWTH University, Aachen)<br /> * Rahil Mahdiyan (Saarland Univ.)<br /> * Hakan Erdogan (Sabanci Univ.)<br /> * Emad Grais (Sabanci Univ.)<br /> * Raghavendra Bilgi (Samsung)<br /> * David Trainor (Sentireal)<br /> * Kenichi Kumatani (Spansion)<br /> * Jordan Cohen (Spelamode)<br /> * Leslie Smith (Univ. Stirling)<br /> * Gérard Chollet (Télécom ParisTech)<br /> * Otman Chakkor (Univ. Tetuan)<br /> * Ngoc Q.K. Duong (Technicolor)<br /> * Zhijian Ou (Tsinghua Univ.)<br /> * Antti Hurmalainen (TU Tampere)<br /> * Tuomas Virtanen (TU Tampere)<br /> * Vinay Modi (Voice Pitara Technologies)<br /> * Jidong Tao (Vlingo)<br /> * Raphaël Blouet</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=List_of_SIG_members&diff=418 List of SIG members 2015-10-10T21:22:55Z <p>Ramon.astudillo: </p> <hr /> <div>118 members as of September 23, 2015:<br /> * Søren Holdt Jensen (Univ. Aalborg)<br /> * Børge Lindberg (Univ. Aalborg)<br /> * Zhen-Hua Tan (Univ. Aalborg)<br /> * Kalle Palomäki (Aalto Univ.)<br /> * Steve Beet (Aculab)<br /> * Gautham Mysore (Adobe)<br /> * Björn Hoffmeister (Amazon)<br /> * Dimitris Dimitriadis (AT&amp;T Research)<br /> * Welly Naptali (ATR-Trek Co.)<br /> * Nassos Katsamanis (TU Athens)<br /> * David Pearce (Audience)<br /> * Wei Xue (Institution of Automation, Chinese Academy of Sciences)<br /> * Keith Ponting (Avaya)<br /> * Daniele Giacobello (Beats by Dr. Dre)<br /> * Armin Sehr (Beuth HS)<br /> * Morgan Nelson (Univ. California, Berkeley)<br /> * Pranav Jawale (IIT Bombay)<br /> * Rajbabu Velmurugan (IIT Bombay)<br /> * Kateřina Žmolíková (Brno University of Technology)<br /> * Mark Gales (Univ. Cambrigde)<br /> * Angel Mario Castro Martínez (Univ. Carl von Ossietzky)<br /> * Bhiksha Raj (Carnegie-Mellon Univ.)<br /> * Rita Singh (Carnegie-Mellon Univ.)<br /> * Richard Stern (Carnegie-Mellon Univ.)<br /> * Climent Nadeu (UP Catalonia)<br /> * Nestor Becerra Yoma (Univ. Chile)<br /> * Fabio Tesser (CNR)<br /> * Jahangir Alam (CRIM)<br /> * Gerhard Taljaard (CSIR)<br /> * John L. Hansen (UT Dallas)<br /> * Taufiq Hasan (UT Dallas)<br /> * Alexey Tarasov (Dublin IT)<br /> * Emanuël Habets (Univ. Erlangen-Nürnberg)<br /> * Walter Kellermann (Univ. Erlangen-Nürnberg)<br /> * Roland Maas (Univ. Erlangen-Nürnberg)<br /> * Georg Stemmer (Univ. Erlangen-Nürnberg)<br /> * Alessio Brutti (FBK)<br /> * Cristina Guerrero Flores (FBK)<br /> * Waheed Mohamed (FBK)<br /> * Brigitte Richardson (Ford Motors)<br /> * Thad Hughes (Google)<br /> * Kevin Wilson (Google)<br /> * José Andrés Gonzalez (Univ. Granada)<br /> * Antonio M. Peinado (Univ. Granada)<br /> * Pejman Mowlaee (Univ. Graz)<br /> * Kwang Myung Jeon (Gwangju IST)<br /> * Martin Heckmann (Honda)<br /> * Afsaneh Asaei (Idiap)<br /> * Ramya Rasipuram (Idiap)<br /> * Alberto Abad (INESC-Id)<br /> * Ramon F. Astudillo (INESC-Id)<br /> * Arseniy Gorin (Inria)<br /> * Juan Andres Morales Cordovilla (Inria)<br /> * Yann Salaün (Inria)<br /> * Sunit Sivasankaran (Inria)<br /> * Dung Tran (Inria)<br /> * Emmanuel Vincent (Inria)<br /> * João Felipe Santos (INRS-EMT)<br /> * Rohit Gowtham (Interactive Intelligence)<br /> * Hynek Hermansky (Johns-Hopkins Univ.)<br /> * Rajesh Hegde (IIT Kanpur)<br /> * Lalan Kumar (IIT Kanpur)<br /> * Karan Nathwani (IIT Kanpur)<br /> * Jort Gemmeke (KU Leuven)<br /> * Hugo van Hamme (KU Leuven)<br /> * Jiri Malek (Univ. Liberec)<br /> * Cong-Thanh Do (LIMSI)<br /> * Rajiv Reddy (LumenVox)<br /> * Alfred Mertins (Univ. Lübeck)<br /> * Florian Müller (Univ. Lübeck)<br /> * Ke Chen (Univ. Manchester)<br /> * Richard Rose (McGill Univ.)<br /> * John Hershey (MERL)<br /> * Jonathan Le Roux (MERL)<br /> * Shinji Watanabe (MERL)<br /> * Mike Seltzer (Microsoft)<br /> * Joseph P. Campbell (MIT)<br /> * Jim Glass (MIT)<br /> * Najim Dehak (MIT)<br /> * Björn Schuller (TU Munich)<br /> * Felix Weninger (TU Munich)<br /> * Lou Boves (Univ. Nijmegen)<br /> * Mario Ganzeboom (Univ. Nijmegen)<br /> * Bert Cranen (Univ. Nijmegen)<br /> * Louis ten Bosch (Univ. Nijmegen)<br /> * David van Leeuwen (Univ. Nijmegen)<br /> * Shuguang Wang (Northwestern Polytechnical Univ., Xi'an)<br /> * Marc Delcroix (NTT)<br /> * Miquel Espi (NTT)<br /> * Keisuke Kinoshita (NTT)<br /> * Cyril Guillaumé (NXP Software)<br /> * Wouter Tirry (NXP Software)<br /> * Arun Narayanan (Ohio State Univ.)<br /> * DeLiang Wang (Ohio State Univ.)<br /> * Simon Bozonnet (OnMobile)<br /> * Anil Alexander (Oxford Wave Research)<br /> * Reinhold Häb-Umbach (Univ. Paderborn)<br /> * Ananya Misra (Princeton Univ.)<br /> * Ricard Marxer (Univ. of Sheffield)<br /> * Ahmed Hussen Abdelaziz (Ruhr Univ., Bochum)<br /> * Dorothea Kolossa (Ruhr Univ., Bochum)<br /> * Tobias Menne (RWTH University, Aachen)<br /> * Rahil Mahdiyan (Saarland Univ.)<br /> * Hakan Erdogan (Sabanci Univ.)<br /> * Emad Grais (Sabanci Univ.)<br /> * Raghavendra Bilgi (Samsung)<br /> * David Trainor (Sentireal)<br /> * Kenichi Kumatani (Spansion)<br /> * Jordan Cohen (Spelamode)<br /> * Leslie Smith (Univ. Stirling)<br /> * Gérard Chollet (Télécom ParisTech)<br /> * Otman Chakkor (Univ. Tetuan)<br /> * Ngoc Q.K. Duong (Technicolor)<br /> * Zhijian Ou (Tsinghua Univ.)<br /> * Antti Hurmalainen (TU Tampere)<br /> * Tuomas Virtanen (TU Tampere)<br /> * Vinay Modi (Voice Pitara Technologies)<br /> * Jidong Tao (Vlingo)<br /> * Raphaël Blouet</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=List_of_SIG_members&diff=415 List of SIG members 2015-09-23T10:28:01Z <p>Ramon.astudillo: </p> <hr /> <div>117 members as of April 7, 2015:<br /> * Søren Holdt Jensen (Univ. Aalborg)<br /> * Børge Lindberg (Univ. Aalborg)<br /> * Zhen-Hua Tan (Univ. Aalborg)<br /> * Kalle Palomäki (Aalto Univ.)<br /> * Steve Beet (Aculab)<br /> * Gautham Mysore (Adobe)<br /> * Björn Hoffmeister (Amazon)<br /> * Dimitris Dimitriadis (AT&amp;T Research)<br /> * Welly Naptali (ATR-Trek Co.)<br /> * Nassos Katsamanis (TU Athens)<br /> * David Pearce (Audience)<br /> * Wei Xue (Institution of Automation, Chinese Academy of Sciences)<br /> * Keith Ponting (Avaya)<br /> * Daniele Giacobello (Beats by Dr. Dre)<br /> * Armin Sehr (Beuth HS)<br /> * Morgan Nelson (Univ. California, Berkeley)<br /> * Pranav Jawale (IIT Bombay)<br /> * Rajbabu Velmurugan (IIT Bombay)<br /> * Mark Gales (Univ. Cambrigde)<br /> * Angel Mario Castro Martínez (Univ. Carl von Ossietzky)<br /> * Bhiksha Raj (Carnegie-Mellon Univ.)<br /> * Rita Singh (Carnegie-Mellon Univ.)<br /> * Richard Stern (Carnegie-Mellon Univ.)<br /> * Climent Nadeu (UP Catalonia)<br /> * Nestor Becerra Yoma (Univ. Chile)<br /> * Fabio Tesser (CNR)<br /> * Jahangir Alam (CRIM)<br /> * Gerhard Taljaard (CSIR)<br /> * John L. Hansen (UT Dallas)<br /> * Taufiq Hasan (UT Dallas)<br /> * Alexey Tarasov (Dublin IT)<br /> * Emanuël Habets (Univ. Erlangen-Nürnberg)<br /> * Walter Kellermann (Univ. Erlangen-Nürnberg)<br /> * Roland Maas (Univ. Erlangen-Nürnberg)<br /> * Georg Stemmer (Univ. Erlangen-Nürnberg)<br /> * Alessio Brutti (FBK)<br /> * Cristina Guerrero Flores (FBK)<br /> * Waheed Mohamed (FBK)<br /> * Brigitte Richardson (Ford Motors)<br /> * Thad Hughes (Google)<br /> * Kevin Wilson (Google)<br /> * José Andrés Gonzalez (Univ. Granada)<br /> * Antonio M. Peinado (Univ. Granada)<br /> * Pejman Mowlaee (Univ. Graz)<br /> * Kwang Myung Jeon (Gwangju IST)<br /> * Martin Heckmann (Honda)<br /> * Afsaneh Asaei (Idiap)<br /> * Ramya Rasipuram (Idiap)<br /> * Alberto Abad (INESC-Id)<br /> * Ramon F. Astudillo (INESC-Id)<br /> * Arseniy Gorin (Inria)<br /> * Juan Andres Morales Cordovilla (Inria)<br /> * Yann Salaün (Inria)<br /> * Sunit Sivasankaran (Inria)<br /> * Dung Tran (Inria)<br /> * Emmanuel Vincent (Inria)<br /> * João Felipe Santos (INRS-EMT)<br /> * Rohit Gowtham (Interactive Intelligence)<br /> * Hynek Hermansky (Johns-Hopkins Univ.)<br /> * Rajesh Hegde (IIT Kanpur)<br /> * Lalan Kumar (IIT Kanpur)<br /> * Karan Nathwani (IIT Kanpur)<br /> * Jort Gemmeke (KU Leuven)<br /> * Hugo van Hamme (KU Leuven)<br /> * Jiri Malek (Univ. Liberec)<br /> * Cong-Thanh Do (LIMSI)<br /> * Rajiv Reddy (LumenVox)<br /> * Alfred Mertins (Univ. Lübeck)<br /> * Florian Müller (Univ. Lübeck)<br /> * Ke Chen (Univ. Manchester)<br /> * Richard Rose (McGill Univ.)<br /> * John Hershey (MERL)<br /> * Jonathan Le Roux (MERL)<br /> * Shinji Watanabe (MERL)<br /> * Mike Seltzer (Microsoft)<br /> * Joseph P. Campbell (MIT)<br /> * Jim Glass (MIT)<br /> * Najim Dehak (MIT)<br /> * Björn Schuller (TU Munich)<br /> * Felix Weninger (TU Munich)<br /> * Lou Boves (Univ. Nijmegen)<br /> * Mario Ganzeboom (Univ. Nijmegen)<br /> * Bert Cranen (Univ. Nijmegen)<br /> * Louis ten Bosch (Univ. Nijmegen)<br /> * David van Leeuwen (Univ. Nijmegen)<br /> * Shuguang Wang (Northwestern Polytechnical Univ., Xi'an)<br /> * Marc Delcroix (NTT)<br /> * Miquel Espi (NTT)<br /> * Keisuke Kinoshita (NTT)<br /> * Cyril Guillaumé (NXP Software)<br /> * Wouter Tirry (NXP Software)<br /> * Arun Narayanan (Ohio State Univ.)<br /> * DeLiang Wang (Ohio State Univ.)<br /> * Simon Bozonnet (OnMobile)<br /> * Anil Alexander (Oxford Wave Research)<br /> * Reinhold Häb-Umbach (Univ. Paderborn)<br /> * Ananya Misra (Princeton Univ.)<br /> * Ricard Marxer (Univ. of Sheffield)<br /> * Ahmed Hussen Abdelaziz (Ruhr Univ., Bochum)<br /> * Dorothea Kolossa (Ruhr Univ., Bochum)<br /> * Tobias Menne (RWTH University, Aachen)<br /> * Rahil Mahdiyan (Saarland Univ.)<br /> * Hakan Erdogan (Sabanci Univ.)<br /> * Emad Grais (Sabanci Univ.)<br /> * Raghavendra Bilgi (Samsung)<br /> * David Trainor (Sentireal)<br /> * Kenichi Kumatani (Spansion)<br /> * Jordan Cohen (Spelamode)<br /> * Leslie Smith (Univ. Stirling)<br /> * Gérard Chollet (Télécom ParisTech)<br /> * Otman Chakkor (Univ. Tetuan)<br /> * Ngoc Q.K. Duong (Technicolor)<br /> * Zhijian Ou (Tsinghua Univ.)<br /> * Antti Hurmalainen (TU Tampere)<br /> * Tuomas Virtanen (TU Tampere)<br /> * Vinay Modi (Voice Pitara Technologies)<br /> * Jidong Tao (Vlingo)<br /> * Raphaël Blouet</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=Software&diff=414 Software 2015-09-19T13:48:10Z <p>Ramon.astudillo: /* Automatic speech recognition */</p> <hr /> <div>This page provides software grouped by application. <br /> <br /> == [[Automatic speech recognition]] ==<br /> {| class=&quot;wikitable sortable&quot; style=&quot;font-size:72%; border:gray solid 1px; text-align:center; width:auto; table-layout:fixed;&quot;<br /> |-<br /> !style=&quot;width: 40px&quot; rowspan=&quot;2&quot; class=&quot;unsortable&quot;| ASR engines<br /> !colspan=&quot;6&quot; |General attributes<br /> !colspan=&quot;2&quot; |Programming<br /> !colspan=&quot;8&quot; |Implemented techniques<br /> !colspan=&quot;2&quot; |Reproducible research<br /> |-<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | release / update<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | actively developed<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | licence<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | platforms<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | links<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | extensions<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | language<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | hardware optimization<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | VAD<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic features<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | feature normalization<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic models<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | model adaptation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | decoding techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | training techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | online ASR<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | robust ASR recipes<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | reproducible results<br /> |-<br /> !CMU Sphinx<br /> |1986-* (Sphinx 4.1.0, pocketsphinx 0.8)<br /> |{{yes|Yes}}<br /> |{{yes|[https://raw.githubusercontent.com/cmusphinx/sphinx4/master/license.terms BSD-like]}}<br /> |{{yes|Windows, Linux, OSX (Sphinx4) / Raspberry-pi, iPhone, Android (pocketsphinx)}}<br /> |[http://cmusphinx.sourceforge.net/ website]<br /> [http://www.cs.cmu.edu/~rsingh/homepage/papers/icassp03-sphinx4_2.pdf paper]<br /> [https://www.cs.cmu.edu/~awb/papers/ICASSP2006/0100185.pdf paper]<br /> [http://sourceforge.net/p/cmusphinx/mailman/ mail-list]<br /> [http://sourceforge.net/p/cmusphinx/discussion/ forum]<br /> [https://github.com/cmusphinx github]<br /> |<br /> |Java (Sphinx4), C (pocketsphinx)<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |CMN, Mel-Spectrum subtraction<br /> |GMM, Streams<br /> |MLLR, MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch<br /> |{{yes|Yes}}<br /> |[http://www.keithv.com/software/sphinx/ AURORA4 (WSJ0)]<br /> |<br /> |-<br /> !HTK<br /> |1993-2009 (3.4.1)<br /> |{{yes|Yes}}<br /> |{{no|[http://htk.eng.cam.ac.uk/docs/license.shtml proprietary]}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://htk.eng.cam.ac.uk/download.shtml website]<br /> [http://htk.eng.cam.ac.uk/docs/docs.shtml book]<br /> [http://htk.eng.cam.ac.uk/mailing/subscribe_mail.shtml mail-list]<br /> |[http://htk.eng.cam.ac.uk/extensions/index.shtml official]<br /> [http://htk.eng.cam.ac.uk/develop/atk.shtml ATK]<br /> [https://github.com/ramon-astudillo/custom_fe uncertain features]<br /> [http://www.astudillo.com/ramon/research/stft-up/ diagonal uncertainty decoding]<br /> [http://full-ud-htk.gforge.inria.fr/ full uncertainty decoding]<br /> |C<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMN<br /> |GMM (Full Cov.), Tied-Mix, Streams<br /> |HLDA, MLLR (w/ reg. trees), CMLR (w/ adaptive training), MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch, MMI, MPE, MWE<br /> |{{yes|Yes}}<br /> |[http://catalog.elra.info/product_info.php?cPath=37_40&amp;products_id=693 AURORA2 (purch.)] [http://catalog.elra.info/index.php?cPath=37_40 AURORA3 (purch.)], [http://www.keithv.com/software/htk/ AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/projects/chime/PCC/data/pcchome.tar.gz CHIME-1], [ftp://ftp.dcs.shef.ac.uk/share/spandh/chime_challenge/grid/eval_tools_grid.tgz CHIME-2-I], [http://spandh.dcs.shef.ac.uk/chime_challenge/chime2_task2.html CHIME-2-II],[http://reverb2014.dereverberation.com/tools/REVERB_TOOLS_FOR_ASR_ver2.0.tgz REVERB]<br /> |ETSI-AFE-AURORA2 [http://aurora.hsnr.de/download/Aurora2_afe_v1_1.pdf paper] (see AURORA2 purch.)<br /> |-<br /> !Kaldi<br /> |2009-* (continous updates)<br /> |{{yes|Yes}}<br /> |{{Yes|Apache 2.0}}<br /> |{{some|Windows (not mantained as of 2014), Linux, OSX}}<br /> |[http://kaldi.sourceforge.net/about.html website]<br /> [http://homepages.inf.ed.ac.uk/aghoshal/pubs/asru11-kaldi.pdf paper]<br /> [http://sourceforge.net/p/kaldi/mailman/kaldi-users/ mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://kaldi.sourceforge.net/install.html SVN]<br /> |[https://github.com/ramon-astudillo/custom_fe uncertain features]<br /> [http://ud-kaldi.gforge.inria.fr/ diagonal uncertainty decoding]<br /> [http://kaldi-to-matlab.gforge.inria.fr/ Matlab conversion tools]<br /> [https://github.com/makladios/Kaldi_Matlab_DNN_UP DNN Uncertainty Decoding]<br /> |C++<br /> |{{yes|BLAS, LAPACK, GPU (for DNNs)}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMVN<br /> |GMM (Full Cov.), SGMM, DNN<br /> |HLDA, STC, MLLT, MLLR, CMLLR (w/ reg. trees), Exponential transform<br /> |aligment, N-best, lattice rescoring (uses OpenFST)<br /> |Baum-Welch, MMI (boosted), MC, feature-based<br /> |{{yes|Yes}}<br /> |[http://kaldi.sourceforge.net/data_prep.html AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/chime_challenge/chime2013/WSJ0public/CHiME2012-WSJ0-Kaldi_0.03.tar.gz CHIME-2]<br /> |Weniger2014-REVERB [http://reverb2014.dereverberation.com/workshop/reverb2014-papers/1569884459.pdf Paper] [http://www.mmk.ei.tum.de/~wen/REVERB_2014/kaldi_baseline.tar.gz Code]<br /> |-<br /> !Spraak<br /> |2008-* (1.1.374)<br /> |{{yes|Yes}}<br /> |{{no|[http://www.spraak.org/obtaining-spraak/license proprietary]}}<br /> |{{some|Windows ([http://www.spraak.org/obtaining-spraak/system-requirements limited]), Linux, OSX}}<br /> |[http://www.spraak.org/ website]<br /> [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/wambacq/interspeech08/is2008_spraak_v3.pdf paper]<br /> [http://www.spraak.org/mailing-lists mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://www.spraak.org/documentation/doxygen/doc/html/spr__svn.html SVN]<br /> |Missing Data Techniques (MDT)<br /> |C, Python<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |Flexible preprocessing script language -- examples for MFCC, PLP<br /> |VTLN,CMN, [http://www.spraak.org/documentation/doxygen/doc/html/spr__tut__mida.html MIDA], [http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__intro.html MDT Techniques], Parametric HistEq [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/ICASSP2010/zhang.pdf&amp;auto&amp;xz:icassp10], Noise normalization [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |GMM (Tied-Mix), Exemplar based [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/dtw/paper_dtw.pdf&amp;auto&amp;kd:icassp11a], NN, CRF, ... (flexible using the preprocessing script) [http://lib.ugent.be/catalog/pug01:4382368]<br /> |CMLLR, eigenvoices, GMM-weight based (NMF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/SPEECOM2013/manuscript_zhang.pdf&amp;auto&amp;xzhang13] -- (all have Matlab dependencies); MAP<br /> |aligment, lattice rescoring, SCRF rescoring (using SCARF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/scarf/paper_scarf.pdf&amp;auto&amp;kd:icassp11b], phone lattice rescoring [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/duchato/eurospeech09/flavor.pdf&amp;auto&amp;jd:intersp09]<br /> |Viterbi<br /> |{{yes|Yes}}<br /> |[http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__example.html AURORA4], [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |-<br /> !Julius<br /> |1997-* (4.3.1)<br /> |{{yes|Yes}}<br /> |{{no|[http://julius.sourceforge.jp/LICENSE.txt propietary]}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://julius.sourceforge.jp/en_index.php website]<br /> [http://jaist.dl.sourceforge.jp/julius/47534/Juliusbook-4.1.5.pdf book]<br /> [http://julius.sourceforge.jp/juliusbook/en/ book online]<br /> [mailto:julius-info@lists.sourceforge.jp mail-list]<br /> [http://julius.sourceforge.jp/forum/ forum]<br /> [http://julius.sourceforge.jp/en_index.php?q=index-en.html#get_by_cvs CVS]<br /> |[http://prdownloads.sourceforge.jp/julius/23332/slf2dfa-1.0.tar.gz htk2Julius grammar]<br /> [http://sourceforge.jp/projects/julius/downloads/32570/julius4-segmentation-kit-v1.0.tar.gz phoneme seg.]<br /> |C<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> | MFCC<br /> |VTLN,CMVN<br /> |GMM (Tied-Mix)<br /> |<br /> |aligment, two-pass decoder<br /> |Baum-Welch<br /> |{{yes|Yes (low latency)}}<br /> |<br /> |<br /> |-<br /> !RWTH<br /> |2001-* (0.6.1)<br /> |{{yes|Yes}}<br /> |{{some|[http://www-i6.informatik.rwth-aachen.de/rwth-asr/rwth-asr-license.html non-commercial]}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://www-i6.informatik.rwth-aachen.de/rwth-asr/ website]<br /> [http://www.isca-speech.org/archive/interspeech_2009/i09_2111.html paper]<br /> [http://www-i6.informatik.rwth-aachen.de/rwth-asr/manual/ wiki]<br /> [http://www-i6.informatik.rwth-aachen.de/rwth-asr/manual/index.php/Special:AWCforum/%3Faction%3Dsc/id3 forum]<br /> |<br /> |C<br /> |{{yes|BLAS, LAPACK, GPU (CUDA), OpenMP}}<br /> |{{yes|Yes}}<br /> | MFCC, PLP, Gammatone, Tandem (MLP)<br /> |VTLN, CMVN, PCA, LDA<br /> |GMM (Tied covariance), DNN<br /> |MLLR, CMLLR, BIC<br /> |aligment, lattice rescoring, system fusion<br /> |Baum-Welch, MPE<br /> |{{yes|Yes}}<br /> |<br /> |}<br /> <br /> == [[Speaker identification and verification]] ==<br /> {| class=&quot;wikitable sortable&quot; style=&quot;font-size:72%; border:gray solid 1px; text-align:center; width:auto; table-layout:fixed;&quot;<br /> |-<br /> !style=&quot;width: 40px&quot; rowspan=&quot;2&quot; class=&quot;unsortable&quot;| Software<br /> !colspan=&quot;6&quot; |General attributes<br /> !colspan=&quot;2&quot; |Programming<br /> !colspan=&quot;8&quot; |Implemented techniques<br /> !colspan=&quot;2&quot; |Reproducible research<br /> |-<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | release / update<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | actively developed<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | licence<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | platforms<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | links<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | extensions<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | language<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | hardware optimization<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | VAD<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic features<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | feature normalization<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | UBM<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | subspace projection<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | subspace normalization<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | scoring<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | diarization<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | robust recognition recipes<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | reproducible results<br /> |-<br /> !BECARS<br /> |2002-2005 (1.1.9)<br /> |{{no|No}}<br /> |{{yes|CeCILL}}<br /> |{{some|Windows, Linux}}<br /> |[http://perso.telecom-paristech.fr/~chollet/becars/ download]<br /> [http://isca-speech.org/archive_open/archive_papers/odyssey_04/ody4_145.pdf paper]<br /> |<br /> |C<br /> |{{no|No}}<br /> |{{no|No}}<br /> |MFCC<br /> |Gaussianization<br /> |GMM<br /> |<br /> |<br /> |MMI-weighted LLR<br /> |{{no|No}}<br /> |<br /> |<br /> |-<br /> !ALIZE<br /> |2005-*<br /> |{{yes|Yes}}<br /> |{{yes|LGPL}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://alize.univ-avignon.fr/ download]<br /> [https://listes.univ-avignon.fr/wws/info/dev-alize mail-list]<br /> [http://www.linkedin.com/groups?mostPopular=&amp;gid=2323703&amp;trk=myg_ugrp_ovr linkedin]<br /> [http://alize.univ-avignon.fr/doc/publi/05_Interspeech_Bonastre.pdf paper]<br /> |<br /> |C++, Perl, Bash<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, LFCC<br /> |CMVN<br /> |GMM<br /> |JFA, i-vector<br /> |whitening, length norm, LDA, WCCN<br /> |cosine, Mahalanobis, SVM, PLDA, Z/T norm<br /> |{{yes|Yes}}<br /> |<br /> |<br /> |-<br /> !LIUM SpkDiarization<br /> |2009-2013<br /> |{{no|No}}<br /> |{{yes|GPL}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://lium3.univ-lemans.fr/diarization/ download]<br /> [http://lium3.univ-lemans.fr/diarization/lib/exe/fetch.php/toolkit-interspeech2013.pdf paper]<br /> |[https://code.google.com/p/voiceid/ Python extension]<br /> |Java<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, LFCC<br /> |CMVN<br /> |GMM<br /> |i-vector<br /> |<br /> |cosine, Mahalanobis<br /> |{{yes|Yes}}<br /> |<br /> |<br /> |-<br /> !MSR Identity Toolbox<br /> |2013<br /> |{{no|No}}<br /> |{{no|[http://research.microsoft.com/en-us/downloads/2476c44a-1f63-4fe0-b805-8c2de395bb2c/MSR-LA%20No%20distrib-OK%20to%20modify.txt proprietary]}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://research.microsoft.com/en-us/downloads/2476c44a-1f63-4fe0-b805-8c2de395bb2c/ download]<br /> [http://research.microsoft.com/pubs/211317/MSR%20Identity%20Toolbox%20v1_1.pdf paper]<br /> |<br /> |Matlab<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC<br /> |CMVN, Gaussianization<br /> |GMM<br /> |i-vector<br /> |whitening, length norm, LDA<br /> |PLDA<br /> |{{no|No}}<br /> |<br /> |<br /> |-<br /> !SPEAR<br /> |2014-*<br /> |{{yes|Yes}}<br /> |{{yes|GPL}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[https://pypi.python.org/pypi/bob.bio.spear download]<br /> [http://publications.idiap.ch/downloads/papers/2014/Khoury_ICASSP_2014.pdf paper]<br /> |<br /> |Python<br /> |{{yes|SGE grid}}<br /> |{{yes|Yes}}<br /> |MFCC, LFCC<br /> |CMVN<br /> |GMM<br /> |ISV, JFA, i-vector<br /> |whitening, length norm, LDA, WCCN<br /> |PLDA, Z/T norm, score fusion<br /> |{{yes|Yes}}<br /> |<br /> |}<br /> <br /> == [[Speech enhancement and separation]] ==<br /> {| class=&quot;wikitable sortable&quot; style=&quot;font-size:72%; border:gray solid 1px; text-align:center; width:auto; table-layout:fixed;&quot;<br /> |-<br /> !style=&quot;width: 40px&quot; rowspan=&quot;2&quot; class=&quot;unsortable&quot;| Software<br /> !colspan=&quot;6&quot; |General attributes<br /> !colspan=&quot;2&quot; |Programming<br /> !colspan=&quot;4&quot; |Implemented techniques<br /> !colspan=&quot;2&quot; |Reproducible research<br /> |-<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | release / update<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | actively developed<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | licence<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | platforms<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | links<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | extensions<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | language<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | hardware optimization<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | spatial model<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | spectral model<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | estimation algorithm<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | online separation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | public recipes<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | reproducible results<br /> |-<br /> !BTK<br /> |2005-*<br /> |{{yes|Yes}}<br /> |{{no|proprietary}}<br /> |{{some|Linux, OSX}}<br /> |[http://distantspeechrecognition.sourceforge.net/ download]<br /> [http://distantspeechrecognition.sourceforge.net/manual.htm#_References papers]<br /> |<br /> |C++, Python<br /> |{{yes|BLAS}}<br /> |DS, SD, MVDR, MN beamforming;<br /> Zelinski, McCowan, Lefkimmiatis post-filters<br /> |none<br /> |GCC-PHAT localization<br /> |{{yes|Yes}}<br /> |<br /> |<br /> |-<br /> !MESSL<br /> |2006-2009<br /> |{{no|No}}<br /> |{{no|proprietary}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[https://github.com/mim/messl download]<br /> [https://www.ee.columbia.edu/~ronw/pubs/taslp09-messl.pdf paper]<br /> |<br /> |Matlab<br /> |{{no|No}}<br /> |IPD/ILD clustering<br /> |none<br /> |EM<br /> |{{no|No}}<br /> |<br /> |<br /> |-<br /> !BeamformIt<br /> |2006-2014 (3.51)<br /> |{{yes|Yes}}<br /> |{{yes|ICSI Open Source Speech Tools}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://www.xavieranguera.com/beamformit/ download]<br /> [http://www.xavieranguera.com/papers/transactions_taslp_2007.pdf paper]<br /> [http://nlp.lsi.upc.edu/papers/thesis_xanguera.pdf thesis]<br /> |<br /> |C++<br /> |{{no|No}}<br /> |weighted DS beamforming<br /> |none<br /> |GCC-PHAT localization<br /> |{{yes|Yes}}<br /> |NIST RT06 (included), [https://github.com/kaldi-asr/kaldi/tree/master/egs/ami/ AMI]<br /> |-<br /> !ManyEars<br /> |2007-2014 (1.1.2)<br /> |{{yes|Yes}}<br /> |{{yes|GPL}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://sourceforge.net/projects/manyears/ download]<br /> [http://link.springer.com/article/10.1007/s10514-012-9316-x# paper]<br /> |<br /> |C<br /> |{{no|No}}<br /> |geometric ICA<br /> |Wiener post-filter (noise only)<br /> |CC-PHAT localization<br /> |{{no|No}}<br /> |<br /> |<br /> |-<br /> !HARK<br /> |2010-* (2.1.2)<br /> |{{yes|Yes}}<br /> |{{some|[http://www.hark.jp/HARK_License_Agreement.pdf non-commercial]}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://www.hark.jp/ download]<br /> [http://www.tandfonline.com/doi/abs/10.1163/016918610X493561#.VeYVmLPcI_s paper]<br /> |<br /> |C++<br /> |{{yes|BLAS}}<br /> |DS, weighted DS, LCMV, GJ, max SNR beamforming;<br /> geometric ICA<br /> |Wiener post-filter (noise only)<br /> |MUSIC localization; MCRA noise estimation<br /> |{{yes|Yes}}<br /> |<br /> |<br /> |-<br /> !FASST<br /> |2012-* (2.0)<br /> |{{yes|Yes}}<br /> |{{yes|QPL}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://bass-db.gforge.inria.fr/fasst/ download]<br /> [https://hal.inria.fr/hal-00626962v2/document paper]<br /> |<br /> |C++, Matlab, Python<br /> |{{yes|OpenMP}}<br /> |full-rank spatial covariance model<br /> |NMF, source-filter NMF, harmonic NMF, smooth NMF<br /> |EM and multiplicative updates<br /> |{{no|No}}<br /> |<br /> |<br /> |}<br /> <br /> == [[Other applications]] ==<br /> <br /> == Contribute software ==<br /> To contribute new software, please<br /> * [[Main_Page#Contribute|create an account]] and login<br /> * go to the wiki page above corresponding to your application; if it does not exist yet, you may create it<br /> * click on the &quot;Edit&quot; link at the top of the page and add a new section for your software (software is ordered by year of the latest version)<br /> * click on the &quot;Save page&quot; link at the bottom of the page to save your modifications<br /> <br /> Please make sure to provide the following information:<br /> * name of the software and year of the latest version<br /> * authors, institution, contact information<br /> * link to the software, ideally including a short demo, and to the external libraries needed<br /> * short description (functionalities, inputs and outputs, programming language, operating system, license, etc) and link to a paper/report describing the software, if any<br /> * whether running on well-known baselines (Aurora-2, Aurora-4, Switchboard, CHiME, etc) is included or requires wrapping by the user<br /> <br /> In order to save storage space, please do not upload the software on this wiki, but link it as much as possible from a public repository (e.g., bitbucket, github, sourceforge) or from a stable URL on the website of your institution. If this is not possible, please contact the [[Main_Page#Working group contacts|resources sharing working group]].</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=List_of_SIG_members&diff=369 List of SIG members 2015-04-07T09:48:19Z <p>Ramon.astudillo: </p> <hr /> <div>116 members as of March 9, 2015:<br /> * Søren Holdt Jensen (Univ. Aalborg)<br /> * Børge Lindberg (Univ. Aalborg)<br /> * Zhen-Hua Tan (Univ. Aalborg)<br /> * Kalle Palomäki (Aalto Univ.)<br /> * Steve Beet (Aculab)<br /> * Gautham Mysore (Adobe)<br /> * Björn Hoffmeister (Amazon)<br /> * Dimitris Dimitriadis (AT&amp;T Research)<br /> * Welly Naptali (ATR-Trek Co.)<br /> * Nassos Katsamanis (TU Athens)<br /> * David Pearce (Audience)<br /> * Wei Xue (Institution of Automation, Chinese Academy of Sciences)<br /> * Keith Ponting (Avaya)<br /> * Daniele Giacobello (Beats by Dr. Dre)<br /> * Armin Sehr (Beuth HS)<br /> * Morgan Nelson (Univ. California, Berkeley)<br /> * Pranav Jawale (IIT Bombay)<br /> * Rajbabu Velmurugan (IIT Bombay)<br /> * Mark Gales (Univ. Cambrigde)<br /> * Angel Mario Castro Martínez (Univ. Carl von Ossietzky)<br /> * Bhiksha Raj (Carnegie-Mellon Univ.)<br /> * Rita Singh (Carnegie-Mellon Univ.)<br /> * Richard Stern (Carnegie-Mellon Univ.)<br /> * Climent Nadeu (UP Catalonia)<br /> * Nestor Becerra Yoma (Univ. Chile)<br /> * Fabio Tesser (CNR)<br /> * Jahangir Alam (CRIM)<br /> * Gerhard Taljaard (CSIR)<br /> * John L. Hansen (UT Dallas)<br /> * Taufiq Hasan (UT Dallas)<br /> * Alexey Tarasov (Dublin IT)<br /> * Emanuël Habets (Univ. Erlangen-Nürnberg)<br /> * Walter Kellermann (Univ. Erlangen-Nürnberg)<br /> * Roland Maas (Univ. Erlangen-Nürnberg)<br /> * Georg Stemmer (Univ. Erlangen-Nürnberg)<br /> * Alessio Brutti (FBK)<br /> * Cristina Guerrero Flores (FBK)<br /> * Waheed Mohamed (FBK)<br /> * Brigitte Richardson (Ford Motors)<br /> * Thad Hughes (Google)<br /> * Kevin Wilson (Google)<br /> * José Andrés Gonzalez (Univ. Granada)<br /> * Antonio M. Peinado (Univ. Granada)<br /> * Pejman Mowlaee (Univ. Graz)<br /> * Kwang Myung Jeon (Gwangju IST)<br /> * Martin Heckmann (Honda)<br /> * Afsaneh Asaei (Idiap)<br /> * Ramya Rasipuram (Idiap)<br /> * Alberto Abad (INESC-Id)<br /> * Ramon F. Astudillo (INESC-Id)<br /> * Arseniy Gorin (Inria)<br /> * Juan Andres Morales Cordovilla (Inria)<br /> * Yann Salaün (Inria)<br /> * Sunit Sivasankaran (Inria)<br /> * Dung Tran (Inria)<br /> * Emmanuel Vincent (Inria)<br /> * João Felipe Santos (INRS-EMT)<br /> * Rohit Gowtham (Interactive Intelligence)<br /> * Hynek Hermansky (Johns-Hopkins Univ.)<br /> * Rajesh Hegde (IIT Kanpur)<br /> * Lalan Kumar (IIT Kanpur)<br /> * Karan Nathwani (IIT Kanpur)<br /> * Jort Gemmeke (KU Leuven)<br /> * Hugo van Hamme (KU Leuven)<br /> * Jiri Malek (Univ. Liberec)<br /> * Cong-Thanh Do (LIMSI)<br /> * Rajiv Reddy (LumenVox)<br /> * Alfred Mertins (Univ. Lübeck)<br /> * Florian Müller (Univ. Lübeck)<br /> * Ke Chen (Univ. Manchester)<br /> * Richard Rose (McGill Univ.)<br /> * John Hershey (MERL)<br /> * Jonathan Le Roux (MERL)<br /> * Shinji Watanabe (MERL)<br /> * Mike Seltzer (Microsoft)<br /> * Joseph P. Campbell (MIT)<br /> * Jim Glass (MIT)<br /> * Najim Dehak (MIT)<br /> * Björn Schuller (TU Munich)<br /> * Felix Weninger (TU Munich)<br /> * Lou Boves (Univ. Nijmegen)<br /> * Mario Ganzeboom (Univ. Nijmegen)<br /> * Bert Cranen (Univ. Nijmegen)<br /> * Louis ten Bosch (Univ. Nijmegen)<br /> * David van Leeuwen (Univ. Nijmegen)<br /> * Shuguang Wang (Northwestern Polytechnical Univ., Xi'an)<br /> * Marc Delcroix (NTT)<br /> * Miquel Espi (NTT)<br /> * Keisuke Kinoshita (NTT)<br /> * Cyril Guillaumé (NXP Software)<br /> * Wouter Tirry (NXP Software)<br /> * Arun Narayanan (Ohio State Univ.)<br /> * DeLiang Wang (Ohio State Univ.)<br /> * Simon Bozonnet (OnMobile)<br /> * Anil Alexander (Oxford Wave Research)<br /> * Reinhold Häb-Umbach (Univ. Paderborn)<br /> * Ananya Misra (Princeton Univ.)<br /> * Ricard Marxer (Univ. of Sheffield)<br /> * Ahmed Hussen Abdelaziz (Ruhr Univ., Bochum)<br /> * Dorothea Kolossa (Ruhr Univ., Bochum)<br /> * Rahil Mahdiyan (Saarland Univ.)<br /> * Hakan Erdogan (Sabanci Univ.)<br /> * Emad Grais (Sabanci Univ.)<br /> * Raghavendra Bilgi (Samsung)<br /> * David Trainor (Sentireal)<br /> * Kenichi Kumatani (Spansion)<br /> * Jordan Cohen (Spelamode)<br /> * Leslie Smith (Univ. Stirling)<br /> * Gérard Chollet (Télécom ParisTech)<br /> * Otman Chakkor (Univ. Tetuan)<br /> * Ngoc Q.K. Duong (Technicolor)<br /> * Zhijian Ou (Tsinghua Univ.)<br /> * Antti Hurmalainen (TU Tampere)<br /> * Tuomas Virtanen (TU Tampere)<br /> * Vinay Modi (Voice Pitara Technologies)<br /> * Jidong Tao (Vlingo)<br /> * Raphaël Blouet</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=List_of_SIG_members&diff=366 List of SIG members 2015-03-09T09:44:22Z <p>Ramon.astudillo: </p> <hr /> <div>113 members as of January 9, 2015:<br /> * Søren Holdt Jensen (Univ. Aalborg)<br /> * Børge Lindberg (Univ. Aalborg)<br /> * Zhen-Hua Tan (Univ. Aalborg)<br /> * Kalle Palomäki (Aalto Univ.)<br /> * Steve Beet (Aculab)<br /> * Gautham Mysore (Adobe)<br /> * Björn Hoffmeister (Amazon)<br /> * Dimitris Dimitriadis (AT&amp;T Research)<br /> * Nassos Katsamanis (TU Athens)<br /> * David Pearce (Audience)<br /> * Wei Xue (Institution of Automation, Chinese Academy of Sciences)<br /> * Keith Ponting (Avaya)<br /> * Daniele Giacobello (Beats by Dr. Dre)<br /> * Armin Sehr (Beuth HS)<br /> * Morgan Nelson (Univ. California, Berkeley)<br /> * Pranav Jawale (IIT Bombay)<br /> * Rajbabu Velmurugan (IIT Bombay)<br /> * Mark Gales (Univ. Cambrigde)<br /> * Angel Mario Castro Martínez (Univ. Carl von Ossietzky)<br /> * Bhiksha Raj (Carnegie-Mellon Univ.)<br /> * Rita Singh (Carnegie-Mellon Univ.)<br /> * Richard Stern (Carnegie-Mellon Univ.)<br /> * Climent Nadeu (UP Catalonia)<br /> * Nestor Becerra Yoma (Univ. Chile)<br /> * Fabio Tesser (CNR)<br /> * Jahangir Alam (CRIM)<br /> * Gerhard Taljaard (CSIR)<br /> * John L. Hansen (UT Dallas)<br /> * Taufiq Hasan (UT Dallas)<br /> * Alexey Tarasov (Dublin IT)<br /> * Emanuël Habets (Univ. Erlangen-Nürnberg)<br /> * Walter Kellermann (Univ. Erlangen-Nürnberg)<br /> * Roland Maas (Univ. Erlangen-Nürnberg)<br /> * Georg Stemmer (Univ. Erlangen-Nürnberg)<br /> * Alessio Brutti (FBK)<br /> * Cristina Guerrero Flores (FBK)<br /> * Waheed Mohamed (FBK)<br /> * Brigitte Richardson (Ford Motors)<br /> * Thad Hughes (Google)<br /> * Kevin Wilson (Google)<br /> * José Andrés Gonzalez (Univ. Granada)<br /> * Antonio M. Peinado (Univ. Granada)<br /> * Pejman Mowlaee (Univ. Graz)<br /> * Kwang Myung Jeon (Gwangju IST)<br /> * Martin Heckmann (Honda)<br /> * Afsaneh Asaei (Idiap)<br /> * Ramya Rasipuram (Idiap)<br /> * Alberto Abad (INESC-Id)<br /> * Ramon F. Astudillo (INESC-Id)<br /> * Arseniy Gorin (Inria)<br /> * Emmanuel Vincent (Inria)<br /> * Dung Tran (Inria)<br /> * Juan Andres Morales Cordovilla (Inria)<br /> * Yann Salaün (Inria)<br /> * João Felipe Santos (INRS-EMT)<br /> * Rohit Gowtham (Interactive Intelligence)<br /> * Hynek Hermansky (Johns-Hopkins Univ.)<br /> * Rajesh Hegde (IIT Kanpur)<br /> * Lalan Kumar (IIT Kanpur)<br /> * Karan Nathwani (IIT Kanpur)<br /> * Jort Gemmeke (KU Leuven)<br /> * Hugo van Hamme (KU Leuven)<br /> * Jiri Malek (Univ. Liberec)<br /> * Cong-Thanh Do (LIMSI)<br /> * Rajiv Reddy (LumenVox)<br /> * Alfred Mertins (Univ. Lübeck)<br /> * Florian Müller (Univ. Lübeck)<br /> * Ke Chen (Univ. Manchester)<br /> * Richard Rose (McGill Univ.)<br /> * John Hershey (MERL)<br /> * Jonathan Le Roux (MERL)<br /> * Shinji Watanabe (MERL)<br /> * Mike Seltzer (Microsoft)<br /> * Joseph P. Campbell (MIT)<br /> * Jim Glass (MIT)<br /> * Najim Dehak (MIT)<br /> * Björn Schuller (TU Munich)<br /> * Felix Weninger (TU Munich)<br /> * Lou Boves (Univ. Nijmegen)<br /> * Mario Ganzeboom (Univ. Nijmegen)<br /> * Bert Cranen (Univ. Nijmegen)<br /> * Louis ten Bosch (Univ. Nijmegen)<br /> * David van Leeuwen (Univ. Nijmegen)<br /> * Shuguang Wang (Northwestern Polytechnical Univ., Xi'an)<br /> * Marc Delcroix (NTT)<br /> * Miquel Espi (NTT)<br /> * Keisuke Kinoshita (NTT)<br /> * Cyril Guillaumé (NXP Software)<br /> * Wouter Tirry (NXP Software)<br /> * Arun Narayanan (Ohio State Univ.)<br /> * DeLiang Wang (Ohio State Univ.)<br /> * Simon Bozonnet (OnMobile)<br /> * Anil Alexander (Oxford Wave Research)<br /> * Reinhold Häb-Umbach (Univ. Paderborn)<br /> * Ananya Misra (Princeton Univ.)<br /> * Ricard Marxer (Univ. of Sheffield)<br /> * Ahmed Hussen Abdelaziz (Ruhr Univ., Bochum)<br /> * Dorothea Kolossa (Ruhr Univ., Bochum)<br /> * Rahil Mahdiyan (Saarland Univ.)<br /> * Hakan Erdogan (Sabanci Univ.)<br /> * Emad Grais (Sabanci Univ.)<br /> * Raghavendra Bilgi (Samsung)<br /> * David Trainor (Sentireal)<br /> * Kenichi Kumatani (Spansion)<br /> * Jordan Cohen (Spelamode)<br /> * Leslie Smith (Univ. Stirling)<br /> * Gérard Chollet (Télécom ParisTech)<br /> * Otman Chakkor (Univ. Tetuan)<br /> * Ngoc Q.K. Duong (Technicolor)<br /> * Zhijian Ou (Tsinghua Univ.)<br /> * Antti Hurmalainen (TU Tampere)<br /> * Tuomas Virtanen (TU Tampere)<br /> * Vinay Modi (Voice Pitara Technologies)<br /> * Jidong Tao (Vlingo)<br /> * Raphaël Blouet</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=Software&diff=365 Software 2015-02-09T18:34:11Z <p>Ramon.astudillo: /* Automatic speech recognition */</p> <hr /> <div>This page provides software grouped by application. <br /> <br /> == [[Automatic speech recognition]] ==<br /> '''Well known ASR engines'''<br /> {| class=&quot;wikitable sortable&quot; style=&quot;font-size:72%; border:gray solid 1px; text-align:center; width:auto; table-layout:fixed;&quot;<br /> |-<br /> !style=&quot;width: 40px&quot; rowspan=&quot;2&quot; class=&quot;unsortable&quot;| ASR engines<br /> !colspan=&quot;6&quot; |General attributes<br /> !colspan=&quot;2&quot; |Programming<br /> !colspan=&quot;8&quot; |Implemented ASR techniques<br /> !colspan=&quot;2&quot; |Reproducible research<br /> |-<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | release / update<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | actively developed<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | licence<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | platforms<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | links<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | extensions<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | language<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | hardware optimization<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | VAD<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic features<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | feature normalization / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic models<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | model adaptation / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | decoding techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | training techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | online ASR<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | robust ASR training recipes<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | reproducible results<br /> |-<br /> !CMU Sphinx<br /> |1986-* (Sphinx 4.1.0, pocketsphinx 0.8)<br /> |{{yes|Yes}}<br /> |{{yes|[https://raw.githubusercontent.com/cmusphinx/sphinx4/master/license.terms BSD-like]}}<br /> |{{yes|Windows, Linux, OSX (Sphinx4) / Raspberry-pi, iPhone, Android (pocketsphinx)}}<br /> |[http://cmusphinx.sourceforge.net/ website]<br /> [http://www.cs.cmu.edu/~rsingh/homepage/papers/icassp03-sphinx4_2.pdf paper]<br /> [https://www.cs.cmu.edu/~awb/papers/ICASSP2006/0100185.pdf paper]<br /> [http://sourceforge.net/p/cmusphinx/mailman/ mail-list]<br /> [http://sourceforge.net/p/cmusphinx/discussion/ forum]<br /> [https://github.com/cmusphinx github]<br /> |<br /> |Java (Sphinx4), C (pocketsphinx)<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |CMN, Mel-Spectrum subtraction<br /> |GMM, Streams<br /> |MLLR, MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch<br /> |{{yes|Yes}}<br /> |[http://www.keithv.com/software/sphinx/ AURORA4 (WSJ0)]<br /> |<br /> |-<br /> !HTK<br /> |1993-2009 (3.4.1)<br /> |{{yes|Yes}}<br /> |{{no|[http://htk.eng.cam.ac.uk/docs/license.shtml proprietary]}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://htk.eng.cam.ac.uk/download.shtml website]<br /> [http://htk.eng.cam.ac.uk/docs/docs.shtml book]<br /> [http://htk.eng.cam.ac.uk/mailing/subscribe_mail.shtml mail-list]<br /> |[http://htk.eng.cam.ac.uk/extensions/index.shtml official]<br /> [http://htk.eng.cam.ac.uk/develop/atk.shtml ATK]<br /> [https://github.com/ramon-astudillo/custom_fe uncertain features]<br /> [http://www.astudillo.com/ramon/research/stft-up/ diagonal uncertainty decoding]<br /> [http://full-ud-htk.gforge.inria.fr/ full uncertainty decoding]<br /> |C<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMN<br /> |GMM (Full Cov.), Tied-Mix, Streams<br /> |HLDA, MLLR (w/ reg. trees), CMLR (w/ adaptive training), MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch, MMI, MPE, MWE<br /> |{{yes|Yes}}<br /> |[http://catalog.elra.info/product_info.php?cPath=37_40&amp;products_id=693 AURORA2 (purch.)] [http://catalog.elra.info/index.php?cPath=37_40 AURORA3 (purch.)], [http://www.keithv.com/software/htk/ AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/projects/chime/PCC/data/pcchome.tar.gz CHIME-1], [ftp://ftp.dcs.shef.ac.uk/share/spandh/chime_challenge/grid/eval_tools_grid.tgz CHIME-2-I], [http://spandh.dcs.shef.ac.uk/chime_challenge/chime2_task2.html CHIME-2-II],[http://reverb2014.dereverberation.com/tools/REVERB_TOOLS_FOR_ASR_ver2.0.tgz REVERB]<br /> |ETSI-AFE-AURORA2 [http://aurora.hsnr.de/download/Aurora2_afe_v1_1.pdf paper] (see AURORA2 purch.)<br /> |-<br /> !Kaldi<br /> |2009-* (continous updates)<br /> |{{yes|Yes}}<br /> |{{Yes|Apache 2.0}}<br /> |{{some|Windows (not mantained as of 2014), Linux, OSX}}<br /> |[http://kaldi.sourceforge.net/about.html website]<br /> [http://homepages.inf.ed.ac.uk/aghoshal/pubs/asru11-kaldi.pdf paper]<br /> [http://sourceforge.net/p/kaldi/mailman/kaldi-users/ mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://kaldi.sourceforge.net/install.html SVN]<br /> |[https://github.com/ramon-astudillo/custom_fe uncertain features]<br /> [http://ud-kaldi.gforge.inria.fr/ diagonal uncertainty decoding]<br /> [http://kaldi-to-matlab.gforge.inria.fr/ Matlab conversion tools]<br /> [https://github.com/makladios/Kaldi_Matlab_DNN_UP DNN Uncertainty Decoding]<br /> |C++<br /> |{{yes|BLAS, LAPACK, GPU (for DNNs)}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMVN<br /> |GMM (Full Cov.), SGMM, DNN<br /> |HLDA, STC, MLLT, MLLR, CMLLR (w/ reg. trees), Exponential transform<br /> |aligment, N-best, lattice rescoring (uses OpenFST)<br /> |Baum-Welch, MMI (boosted), MC, feature-based<br /> |{{yes|Yes}}<br /> |[http://kaldi.sourceforge.net/data_prep.html AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/chime_challenge/WSJ0public/CHiME2012-WSJ0-Kaldi_0.03.tar.gz CHIME-2]<br /> |Weniger2014-REVERB [http://reverb2014.dereverberation.com/workshop/reverb2014-papers/1569884459.pdf Paper] [http://www.mmk.ei.tum.de/~wen/REVERB_2014/kaldi_baseline.tar.gz Code]<br /> |-<br /> !Spraak<br /> |2008-* (1.1.374)<br /> |{{yes|Yes}}<br /> |{{no|[http://www.spraak.org/obtaining-spraak/license proprietary]}}<br /> |{{some|Windows ([http://www.spraak.org/obtaining-spraak/system-requirements limited]), Linux, OSX}}<br /> |[http://www.spraak.org/ website]<br /> [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/wambacq/interspeech08/is2008_spraak_v3.pdf paper]<br /> [http://www.spraak.org/mailing-lists mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://www.spraak.org/documentation/doxygen/doc/html/spr__svn.html SVN]<br /> |Missing Data Techniques (MDT)<br /> |C, Python<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |Flexible preprocessing script language -- examples for MFCC, PLP<br /> |VTLN,CMN, [http://www.spraak.org/documentation/doxygen/doc/html/spr__tut__mida.html MIDA], [http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__intro.html MDT Techniques], Parametric HistEq [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/ICASSP2010/zhang.pdf&amp;auto&amp;xz:icassp10], Noise normalization [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |GMM (Tied-Mix), Exemplar based [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/dtw/paper_dtw.pdf&amp;auto&amp;kd:icassp11a], NN, CRF, ... (flexible using the preprocessing script) [http://lib.ugent.be/catalog/pug01:4382368]<br /> |CMLLR, eigenvoices, GMM-weight based (NMF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/SPEECOM2013/manuscript_zhang.pdf&amp;auto&amp;xzhang13] -- (all have Matlab dependencies); MAP<br /> |aligment, lattice rescoring, SCRF rescoring (using SCARF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/scarf/paper_scarf.pdf&amp;auto&amp;kd:icassp11b], phone lattice rescoring [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/duchato/eurospeech09/flavor.pdf&amp;auto&amp;jd:intersp09]<br /> |Viterbi<br /> |{{yes|Yes}}<br /> |[http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__example.html AURORA4], [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |-<br /> !Julius<br /> |1997-* (4.3.1)<br /> |{{yes|Yes}}<br /> |{{no|[http://julius.sourceforge.jp/LICENSE.txt propietary]}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://julius.sourceforge.jp/en_index.php website]<br /> [http://jaist.dl.sourceforge.jp/julius/47534/Juliusbook-4.1.5.pdf book]<br /> [http://julius.sourceforge.jp/juliusbook/en/ book online]<br /> [mailto:julius-info@lists.sourceforge.jp mail-list]<br /> [http://julius.sourceforge.jp/forum/ forum]<br /> [http://julius.sourceforge.jp/en_index.php?q=index-en.html#get_by_cvs CVS]<br /> |[http://prdownloads.sourceforge.jp/julius/23332/slf2dfa-1.0.tar.gz htk2Julius grammar]<br /> [http://sourceforge.jp/projects/julius/downloads/32570/julius4-segmentation-kit-v1.0.tar.gz phoneme seg.]<br /> |C<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> | MFCC<br /> |VTLN,CMVN<br /> |GMM (Tied-Mix)<br /> |<br /> |aligment, two-pass decoder<br /> |Baum-Welch<br /> |{{yes|Yes (low latency)}}<br /> |<br /> |<br /> |-<br /> !RWTH<br /> |2001-* (0.6.1)<br /> |{{yes|Yes}}<br /> |{{some|[http://www-i6.informatik.rwth-aachen.de/rwth-asr/rwth-asr-license.html non-commercial]}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://www-i6.informatik.rwth-aachen.de/rwth-asr/ website]<br /> [http://www.isca-speech.org/archive/interspeech_2009/i09_2111.html paper]<br /> [http://www-i6.informatik.rwth-aachen.de/rwth-asr/manual/ wiki]<br /> [http://www-i6.informatik.rwth-aachen.de/rwth-asr/manual/index.php/Special:AWCforum/%3Faction%3Dsc/id3 forum]<br /> |<br /> |C<br /> |{{yes|BLAS, LAPACK, GPU (CUDA), OpenMP}}<br /> |{{yes|Yes}}<br /> | MFCC, PLP, Gammatone, Tandem (MLP)<br /> |VTLN, CMVN, PCA, LDA<br /> |GMM (Tied covariance), DNN<br /> |MLLR, CMLLR, BIC<br /> |aligment, lattice rescoring, system fusion<br /> |Baum-Welch, MPE<br /> |{{yes|Yes}}<br /> |<br /> |}<br /> '''Other Software'''<br /> <br /> == [[Speaker identification and verification]] ==<br /> <br /> == [[Speech enhancement and separation]] ==<br /> <br /> == [[Other applications]] ==<br /> <br /> == Contribute software ==<br /> To contribute new software, please<br /> * [[Special:UserLogin|create an account]] and login<br /> * go to the wiki page above corresponding to your application; if it does not exist yet, you may create it<br /> * click on the &quot;Edit&quot; link at the top of the page and add a new section for your software (software is ordered by year of the latest version)<br /> * click on the &quot;Save page&quot; link at the bottom of the page to save your modifications<br /> <br /> Please make sure to provide the following information:<br /> * name of the software and year of the latest version<br /> * authors, institution, contact information<br /> * link to the software, ideally including a short demo, and to the external libraries needed<br /> * short description (functionalities, inputs and outputs, programming language, operating system, license, etc) and link to a paper/report describing the software, if any<br /> * whether running on well-known baselines (Aurora-2, Aurora-4, Switchboard, CHiME, etc) is included or requires wrapping by the user<br /> <br /> In order to save storage space, please do not upload the software on this wiki, but link it as much as possible from a public repository (e.g., bitbucket, github, sourceforge) or from a stable URL on the website of your institution. If this is not possible, please contact the [[Main_Page#Working group contacts|resources sharing working group]].</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=Software&diff=364 Software 2015-02-09T18:30:53Z <p>Ramon.astudillo: /* Automatic speech recognition */</p> <hr /> <div>This page provides software grouped by application. <br /> <br /> == [[Automatic speech recognition]] ==<br /> '''Well known ASR engines'''<br /> {| class=&quot;wikitable sortable&quot; style=&quot;font-size:72%; border:gray solid 1px; text-align:center; width:auto; table-layout:fixed;&quot;<br /> |-<br /> !style=&quot;width: 40px&quot; rowspan=&quot;2&quot; class=&quot;unsortable&quot;| ASR engines<br /> !colspan=&quot;6&quot; |General attributes<br /> !colspan=&quot;2&quot; |Programming<br /> !colspan=&quot;8&quot; |Implemented ASR techniques<br /> !colspan=&quot;2&quot; |Reproducible research<br /> |-<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | release / update<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | actively developed<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | licence<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | platforms<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | links<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | extensions<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | language<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | hardware optimization<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | VAD<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic features<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | feature normalization / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic models<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | model adaptation / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | decoding techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | training techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | online ASR<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | robust ASR training recipes<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | reproducible results<br /> |-<br /> !CMU Sphinx<br /> |1986-* (Sphinx 4.1.0, pocketsphinx 0.8)<br /> |{{yes|Yes}}<br /> |{{yes|[https://raw.githubusercontent.com/cmusphinx/sphinx4/master/license.terms BSD-like]}}<br /> |{{yes|Windows, Linux, OSX (Sphinx4) / Raspberry-pi, iPhone, Android (pocketsphinx)}}<br /> |[http://cmusphinx.sourceforge.net/ website]<br /> [http://www.cs.cmu.edu/~rsingh/homepage/papers/icassp03-sphinx4_2.pdf paper]<br /> [https://www.cs.cmu.edu/~awb/papers/ICASSP2006/0100185.pdf paper]<br /> [http://sourceforge.net/p/cmusphinx/mailman/ mail-list]<br /> [http://sourceforge.net/p/cmusphinx/discussion/ forum]<br /> [https://github.com/cmusphinx github]<br /> |<br /> |Java (Sphinx4), C (pocketsphinx)<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |CMN, Mel-Spectrum subtraction<br /> |GMM, Streams<br /> |MLLR, MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch<br /> |{{yes|Yes}}<br /> |[http://www.keithv.com/software/sphinx/ AURORA4 (WSJ0)]<br /> |<br /> |-<br /> !HTK<br /> |1993-2009 (3.4.1)<br /> |{{yes|Yes}}<br /> |{{no|[http://htk.eng.cam.ac.uk/docs/license.shtml proprietary]}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://htk.eng.cam.ac.uk/download.shtml website]<br /> [http://htk.eng.cam.ac.uk/docs/docs.shtml book]<br /> [http://htk.eng.cam.ac.uk/mailing/subscribe_mail.shtml mail-list]<br /> |[http://htk.eng.cam.ac.uk/extensions/index.shtml official]<br /> [http://htk.eng.cam.ac.uk/develop/atk.shtml ATK]<br /> [https://github.com/ramon-astudillo/custom_fe diagonal uncertainty decoding]<br /> [http://full-ud-htk.gforge.inria.fr/ full uncertainty decoding]<br /> |C<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMN<br /> |GMM (Full Cov.), Tied-Mix, Streams<br /> |HLDA, MLLR (w/ reg. trees), CMLR (w/ adaptive training), MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch, MMI, MPE, MWE<br /> |{{yes|Yes}}<br /> |[http://catalog.elra.info/product_info.php?cPath=37_40&amp;products_id=693 AURORA2 (purch.)] [http://catalog.elra.info/index.php?cPath=37_40 AURORA3 (purch.)], [http://www.keithv.com/software/htk/ AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/projects/chime/PCC/data/pcchome.tar.gz CHIME-1], [ftp://ftp.dcs.shef.ac.uk/share/spandh/chime_challenge/grid/eval_tools_grid.tgz CHIME-2-I], [http://spandh.dcs.shef.ac.uk/chime_challenge/chime2_task2.html CHIME-2-II],[http://reverb2014.dereverberation.com/tools/REVERB_TOOLS_FOR_ASR_ver2.0.tgz REVERB]<br /> |ETSI-AFE-AURORA2 [http://aurora.hsnr.de/download/Aurora2_afe_v1_1.pdf paper] (see AURORA2 purch.)<br /> |-<br /> !Kaldi<br /> |2009-* (continous updates)<br /> |{{yes|Yes}}<br /> |{{Yes|Apache 2.0}}<br /> |{{some|Windows (not mantained as of 2014), Linux, OSX}}<br /> |[http://kaldi.sourceforge.net/about.html website]<br /> [http://homepages.inf.ed.ac.uk/aghoshal/pubs/asru11-kaldi.pdf paper]<br /> [http://sourceforge.net/p/kaldi/mailman/kaldi-users/ mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://kaldi.sourceforge.net/install.html SVN]<br /> |[http://ud-kaldi.gforge.inria.fr/ diagonal uncertainty decoding]<br /> [http://kaldi-to-matlab.gforge.inria.fr/ Matlab conversion tools]<br /> [https://github.com/makladios/Kaldi_Matlab_DNN_UP DNN Uncertainty Decoding]<br /> |C++<br /> |{{yes|BLAS, LAPACK, GPU (for DNNs)}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMVN<br /> |GMM (Full Cov.), SGMM, DNN<br /> |HLDA, STC, MLLT, MLLR, CMLLR (w/ reg. trees), Exponential transform<br /> |aligment, N-best, lattice rescoring (uses OpenFST)<br /> |Baum-Welch, MMI (boosted), MC, feature-based<br /> |{{yes|Yes}}<br /> |[http://kaldi.sourceforge.net/data_prep.html AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/chime_challenge/WSJ0public/CHiME2012-WSJ0-Kaldi_0.03.tar.gz CHIME-2]<br /> |Weniger2014-REVERB [http://reverb2014.dereverberation.com/workshop/reverb2014-papers/1569884459.pdf Paper] [http://www.mmk.ei.tum.de/~wen/REVERB_2014/kaldi_baseline.tar.gz Code]<br /> |-<br /> !Spraak<br /> |2008-* (1.1.374)<br /> |{{yes|Yes}}<br /> |{{no|[http://www.spraak.org/obtaining-spraak/license proprietary]}}<br /> |{{some|Windows ([http://www.spraak.org/obtaining-spraak/system-requirements limited]), Linux, OSX}}<br /> |[http://www.spraak.org/ website]<br /> [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/wambacq/interspeech08/is2008_spraak_v3.pdf paper]<br /> [http://www.spraak.org/mailing-lists mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://www.spraak.org/documentation/doxygen/doc/html/spr__svn.html SVN]<br /> |Missing Data Techniques (MDT)<br /> |C, Python<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |Flexible preprocessing script language -- examples for MFCC, PLP<br /> |VTLN,CMN, [http://www.spraak.org/documentation/doxygen/doc/html/spr__tut__mida.html MIDA], [http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__intro.html MDT Techniques], Parametric HistEq [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/ICASSP2010/zhang.pdf&amp;auto&amp;xz:icassp10], Noise normalization [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |GMM (Tied-Mix), Exemplar based [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/dtw/paper_dtw.pdf&amp;auto&amp;kd:icassp11a], NN, CRF, ... (flexible using the preprocessing script) [http://lib.ugent.be/catalog/pug01:4382368]<br /> |CMLLR, eigenvoices, GMM-weight based (NMF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/SPEECOM2013/manuscript_zhang.pdf&amp;auto&amp;xzhang13] -- (all have Matlab dependencies); MAP<br /> |aligment, lattice rescoring, SCRF rescoring (using SCARF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/scarf/paper_scarf.pdf&amp;auto&amp;kd:icassp11b], phone lattice rescoring [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/duchato/eurospeech09/flavor.pdf&amp;auto&amp;jd:intersp09]<br /> |Viterbi<br /> |{{yes|Yes}}<br /> |[http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__example.html AURORA4], [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |-<br /> !Julius<br /> |1997-* (4.3.1)<br /> |{{yes|Yes}}<br /> |{{no|[http://julius.sourceforge.jp/LICENSE.txt propietary]}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://julius.sourceforge.jp/en_index.php website]<br /> [http://jaist.dl.sourceforge.jp/julius/47534/Juliusbook-4.1.5.pdf book]<br /> [http://julius.sourceforge.jp/juliusbook/en/ book online]<br /> [mailto:julius-info@lists.sourceforge.jp mail-list]<br /> [http://julius.sourceforge.jp/forum/ forum]<br /> [http://julius.sourceforge.jp/en_index.php?q=index-en.html#get_by_cvs CVS]<br /> |[http://prdownloads.sourceforge.jp/julius/23332/slf2dfa-1.0.tar.gz htk2Julius grammar]<br /> [http://sourceforge.jp/projects/julius/downloads/32570/julius4-segmentation-kit-v1.0.tar.gz phoneme seg.]<br /> |C<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> | MFCC<br /> |VTLN,CMVN<br /> |GMM (Tied-Mix)<br /> |<br /> |aligment, two-pass decoder<br /> |Baum-Welch<br /> |{{yes|Yes (low latency)}}<br /> |<br /> |<br /> |-<br /> !RWTH<br /> |2001-* (0.6.1)<br /> |{{yes|Yes}}<br /> |{{some|[http://www-i6.informatik.rwth-aachen.de/rwth-asr/rwth-asr-license.html non-commercial]}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://www-i6.informatik.rwth-aachen.de/rwth-asr/ website]<br /> [http://www.isca-speech.org/archive/interspeech_2009/i09_2111.html paper]<br /> [http://www-i6.informatik.rwth-aachen.de/rwth-asr/manual/ wiki]<br /> [http://www-i6.informatik.rwth-aachen.de/rwth-asr/manual/index.php/Special:AWCforum/%3Faction%3Dsc/id3 forum]<br /> |<br /> |C<br /> |{{yes|BLAS, LAPACK, GPU (CUDA), OpenMP}}<br /> |{{yes|Yes}}<br /> | MFCC, PLP, Gammatone, Tandem (MLP)<br /> |VTLN, CMVN, PCA, LDA<br /> |GMM (Tied covariance), DNN<br /> |MLLR, CMLLR, BIC<br /> |aligment, lattice rescoring, system fusion<br /> |Baum-Welch, MPE<br /> |{{yes|Yes}}<br /> |<br /> |}<br /> '''Other Software'''<br /> <br /> == [[Speaker identification and verification]] ==<br /> <br /> == [[Speech enhancement and separation]] ==<br /> <br /> == [[Other applications]] ==<br /> <br /> == Contribute software ==<br /> To contribute new software, please<br /> * [[Special:UserLogin|create an account]] and login<br /> * go to the wiki page above corresponding to your application; if it does not exist yet, you may create it<br /> * click on the &quot;Edit&quot; link at the top of the page and add a new section for your software (software is ordered by year of the latest version)<br /> * click on the &quot;Save page&quot; link at the bottom of the page to save your modifications<br /> <br /> Please make sure to provide the following information:<br /> * name of the software and year of the latest version<br /> * authors, institution, contact information<br /> * link to the software, ideally including a short demo, and to the external libraries needed<br /> * short description (functionalities, inputs and outputs, programming language, operating system, license, etc) and link to a paper/report describing the software, if any<br /> * whether running on well-known baselines (Aurora-2, Aurora-4, Switchboard, CHiME, etc) is included or requires wrapping by the user<br /> <br /> In order to save storage space, please do not upload the software on this wiki, but link it as much as possible from a public repository (e.g., bitbucket, github, sourceforge) or from a stable URL on the website of your institution. If this is not possible, please contact the [[Main_Page#Working group contacts|resources sharing working group]].</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=Software&diff=363 Software 2015-02-09T18:29:52Z <p>Ramon.astudillo: /* Automatic speech recognition */</p> <hr /> <div>This page provides software grouped by application. <br /> <br /> == [[Automatic speech recognition]] ==<br /> '''Well known ASR engines'''<br /> {| class=&quot;wikitable sortable&quot; style=&quot;font-size:72%; border:gray solid 1px; text-align:center; width:auto; table-layout:fixed;&quot;<br /> |-<br /> !style=&quot;width: 40px&quot; rowspan=&quot;2&quot; class=&quot;unsortable&quot;| ASR engines<br /> !colspan=&quot;6&quot; |General attributes<br /> !colspan=&quot;2&quot; |Programming<br /> !colspan=&quot;8&quot; |Implemented ASR techniques<br /> !colspan=&quot;2&quot; |Reproducible research<br /> |-<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | release / update<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | actively developed<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | licence<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | platforms<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | links<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | extensions<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | language<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | hardware optimization<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | VAD<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic features<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | feature normalization / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic models<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | model adaptation / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | decoding techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | training techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | online ASR<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | robust ASR training recipes<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | reproducible results<br /> |-<br /> !CMU Sphinx<br /> |1986-* (Sphinx 4.1.0, pocketsphinx 0.8)<br /> |{{yes|Yes}}<br /> |{{yes|[https://raw.githubusercontent.com/cmusphinx/sphinx4/master/license.terms BSD-like]}}<br /> |{{yes|Windows, Linux, OSX (Sphinx4) / Raspberry-pi, iPhone, Android (pocketsphinx)}}<br /> |[http://cmusphinx.sourceforge.net/ website]<br /> [http://www.cs.cmu.edu/~rsingh/homepage/papers/icassp03-sphinx4_2.pdf paper]<br /> [https://www.cs.cmu.edu/~awb/papers/ICASSP2006/0100185.pdf paper]<br /> [http://sourceforge.net/p/cmusphinx/mailman/ mail-list]<br /> [http://sourceforge.net/p/cmusphinx/discussion/ forum]<br /> [https://github.com/cmusphinx github]<br /> |<br /> |Java (Sphinx4), C (pocketsphinx)<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |CMN, Mel-Spectrum subtraction<br /> |GMM, Streams<br /> |MLLR, MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch<br /> |{{yes|Yes}}<br /> |[http://www.keithv.com/software/sphinx/ AURORA4 (WSJ0)]<br /> |<br /> |-<br /> !HTK<br /> |1993-2009 (3.4.1)<br /> |{{yes|Yes}}<br /> |{{no|[http://htk.eng.cam.ac.uk/docs/license.shtml proprietary]}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://htk.eng.cam.ac.uk/download.shtml website]<br /> [http://htk.eng.cam.ac.uk/docs/docs.shtml book]<br /> [http://htk.eng.cam.ac.uk/mailing/subscribe_mail.shtml mail-list]<br /> |[http://htk.eng.cam.ac.uk/extensions/index.shtml official]<br /> [http://htk.eng.cam.ac.uk/develop/atk.shtml ATK]<br /> [https://github.com/ramon-astudillo/custom_fe diagonal uncertainty decoding]<br /> [https://github.com/makladios/Kaldi_Matlab_DNN_UP DNN Uncertainty Decoding]<br /> [http://full-ud-htk.gforge.inria.fr/ full uncertainty decoding]<br /> |C<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMN<br /> |GMM (Full Cov.), Tied-Mix, Streams<br /> |HLDA, MLLR (w/ reg. trees), CMLR (w/ adaptive training), MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch, MMI, MPE, MWE<br /> |{{yes|Yes}}<br /> |[http://catalog.elra.info/product_info.php?cPath=37_40&amp;products_id=693 AURORA2 (purch.)] [http://catalog.elra.info/index.php?cPath=37_40 AURORA3 (purch.)], [http://www.keithv.com/software/htk/ AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/projects/chime/PCC/data/pcchome.tar.gz CHIME-1], [ftp://ftp.dcs.shef.ac.uk/share/spandh/chime_challenge/grid/eval_tools_grid.tgz CHIME-2-I], [http://spandh.dcs.shef.ac.uk/chime_challenge/chime2_task2.html CHIME-2-II],[http://reverb2014.dereverberation.com/tools/REVERB_TOOLS_FOR_ASR_ver2.0.tgz REVERB]<br /> |ETSI-AFE-AURORA2 [http://aurora.hsnr.de/download/Aurora2_afe_v1_1.pdf paper] (see AURORA2 purch.)<br /> |-<br /> !Kaldi<br /> |2009-* (continous updates)<br /> |{{yes|Yes}}<br /> |{{Yes|Apache 2.0}}<br /> |{{some|Windows (not mantained as of 2014), Linux, OSX}}<br /> |[http://kaldi.sourceforge.net/about.html website]<br /> [http://homepages.inf.ed.ac.uk/aghoshal/pubs/asru11-kaldi.pdf paper]<br /> [http://sourceforge.net/p/kaldi/mailman/kaldi-users/ mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://kaldi.sourceforge.net/install.html SVN]<br /> |[http://ud-kaldi.gforge.inria.fr/ diagonal uncertainty decoding]<br /> [http://kaldi-to-matlab.gforge.inria.fr/ Matlab conversion tools]<br /> |C++<br /> |{{yes|BLAS, LAPACK, GPU (for DNNs)}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMVN<br /> |GMM (Full Cov.), SGMM, DNN<br /> |HLDA, STC, MLLT, MLLR, CMLLR (w/ reg. trees), Exponential transform<br /> |aligment, N-best, lattice rescoring (uses OpenFST)<br /> |Baum-Welch, MMI (boosted), MC, feature-based<br /> |{{yes|Yes}}<br /> |[http://kaldi.sourceforge.net/data_prep.html AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/chime_challenge/WSJ0public/CHiME2012-WSJ0-Kaldi_0.03.tar.gz CHIME-2]<br /> |Weniger2014-REVERB [http://reverb2014.dereverberation.com/workshop/reverb2014-papers/1569884459.pdf Paper] [http://www.mmk.ei.tum.de/~wen/REVERB_2014/kaldi_baseline.tar.gz Code]<br /> |-<br /> !Spraak<br /> |2008-* (1.1.374)<br /> |{{yes|Yes}}<br /> |{{no|[http://www.spraak.org/obtaining-spraak/license proprietary]}}<br /> |{{some|Windows ([http://www.spraak.org/obtaining-spraak/system-requirements limited]), Linux, OSX}}<br /> |[http://www.spraak.org/ website]<br /> [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/wambacq/interspeech08/is2008_spraak_v3.pdf paper]<br /> [http://www.spraak.org/mailing-lists mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://www.spraak.org/documentation/doxygen/doc/html/spr__svn.html SVN]<br /> |Missing Data Techniques (MDT)<br /> |C, Python<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |Flexible preprocessing script language -- examples for MFCC, PLP<br /> |VTLN,CMN, [http://www.spraak.org/documentation/doxygen/doc/html/spr__tut__mida.html MIDA], [http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__intro.html MDT Techniques], Parametric HistEq [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/ICASSP2010/zhang.pdf&amp;auto&amp;xz:icassp10], Noise normalization [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |GMM (Tied-Mix), Exemplar based [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/dtw/paper_dtw.pdf&amp;auto&amp;kd:icassp11a], NN, CRF, ... (flexible using the preprocessing script) [http://lib.ugent.be/catalog/pug01:4382368]<br /> |CMLLR, eigenvoices, GMM-weight based (NMF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/SPEECOM2013/manuscript_zhang.pdf&amp;auto&amp;xzhang13] -- (all have Matlab dependencies); MAP<br /> |aligment, lattice rescoring, SCRF rescoring (using SCARF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/scarf/paper_scarf.pdf&amp;auto&amp;kd:icassp11b], phone lattice rescoring [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/duchato/eurospeech09/flavor.pdf&amp;auto&amp;jd:intersp09]<br /> |Viterbi<br /> |{{yes|Yes}}<br /> |[http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__example.html AURORA4], [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |-<br /> !Julius<br /> |1997-* (4.3.1)<br /> |{{yes|Yes}}<br /> |{{no|[http://julius.sourceforge.jp/LICENSE.txt propietary]}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://julius.sourceforge.jp/en_index.php website]<br /> [http://jaist.dl.sourceforge.jp/julius/47534/Juliusbook-4.1.5.pdf book]<br /> [http://julius.sourceforge.jp/juliusbook/en/ book online]<br /> [mailto:julius-info@lists.sourceforge.jp mail-list]<br /> [http://julius.sourceforge.jp/forum/ forum]<br /> [http://julius.sourceforge.jp/en_index.php?q=index-en.html#get_by_cvs CVS]<br /> |[http://prdownloads.sourceforge.jp/julius/23332/slf2dfa-1.0.tar.gz htk2Julius grammar]<br /> [http://sourceforge.jp/projects/julius/downloads/32570/julius4-segmentation-kit-v1.0.tar.gz phoneme seg.]<br /> |C<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> | MFCC<br /> |VTLN,CMVN<br /> |GMM (Tied-Mix)<br /> |<br /> |aligment, two-pass decoder<br /> |Baum-Welch<br /> |{{yes|Yes (low latency)}}<br /> |<br /> |<br /> |-<br /> !RWTH<br /> |2001-* (0.6.1)<br /> |{{yes|Yes}}<br /> |{{some|[http://www-i6.informatik.rwth-aachen.de/rwth-asr/rwth-asr-license.html non-commercial]}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://www-i6.informatik.rwth-aachen.de/rwth-asr/ website]<br /> [http://www.isca-speech.org/archive/interspeech_2009/i09_2111.html paper]<br /> [http://www-i6.informatik.rwth-aachen.de/rwth-asr/manual/ wiki]<br /> [http://www-i6.informatik.rwth-aachen.de/rwth-asr/manual/index.php/Special:AWCforum/%3Faction%3Dsc/id3 forum]<br /> |<br /> |C<br /> |{{yes|BLAS, LAPACK, GPU (CUDA), OpenMP}}<br /> |{{yes|Yes}}<br /> | MFCC, PLP, Gammatone, Tandem (MLP)<br /> |VTLN, CMVN, PCA, LDA<br /> |GMM (Tied covariance), DNN<br /> |MLLR, CMLLR, BIC<br /> |aligment, lattice rescoring, system fusion<br /> |Baum-Welch, MPE<br /> |{{yes|Yes}}<br /> |<br /> |}<br /> '''Other Software'''<br /> <br /> == [[Speaker identification and verification]] ==<br /> <br /> == [[Speech enhancement and separation]] ==<br /> <br /> == [[Other applications]] ==<br /> <br /> == Contribute software ==<br /> To contribute new software, please<br /> * [[Special:UserLogin|create an account]] and login<br /> * go to the wiki page above corresponding to your application; if it does not exist yet, you may create it<br /> * click on the &quot;Edit&quot; link at the top of the page and add a new section for your software (software is ordered by year of the latest version)<br /> * click on the &quot;Save page&quot; link at the bottom of the page to save your modifications<br /> <br /> Please make sure to provide the following information:<br /> * name of the software and year of the latest version<br /> * authors, institution, contact information<br /> * link to the software, ideally including a short demo, and to the external libraries needed<br /> * short description (functionalities, inputs and outputs, programming language, operating system, license, etc) and link to a paper/report describing the software, if any<br /> * whether running on well-known baselines (Aurora-2, Aurora-4, Switchboard, CHiME, etc) is included or requires wrapping by the user<br /> <br /> In order to save storage space, please do not upload the software on this wiki, but link it as much as possible from a public repository (e.g., bitbucket, github, sourceforge) or from a stable URL on the website of your institution. If this is not possible, please contact the [[Main_Page#Working group contacts|resources sharing working group]].</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=Software&diff=355 Software 2015-01-13T14:51:43Z <p>Ramon.astudillo: /* Automatic speech recognition */</p> <hr /> <div>This page provides software grouped by application. <br /> <br /> == [[Automatic speech recognition]] ==<br /> '''Well known ASR engines'''<br /> {| class=&quot;wikitable sortable&quot; style=&quot;font-size:72%; border:gray solid 1px; text-align:center; width:auto; table-layout:fixed;&quot;<br /> |-<br /> !style=&quot;width: 40px&quot; rowspan=&quot;2&quot; class=&quot;unsortable&quot;| ASR engines<br /> !colspan=&quot;6&quot; |General attributes<br /> !colspan=&quot;2&quot; |Programming<br /> !colspan=&quot;8&quot; |Implemented ASR techniques<br /> !colspan=&quot;2&quot; |Reproducible research<br /> |-<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | release / update<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | actively developed<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | licence<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | platforms<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | links<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | extensions<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | language<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | hardware optimization<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | VAD<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic features<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | feature normalization / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic models<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | model adaptation / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | decoding techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | training techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | online ASR<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | robust ASR training recipes<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | reproducible results<br /> |-<br /> !CMU Sphinx<br /> |1986-* (Sphinx 4.1.0, pocketsphinx 0.8)<br /> |{{yes|Yes}}<br /> |{{yes|[https://raw.githubusercontent.com/cmusphinx/sphinx4/master/license.terms BSD-like]}}<br /> |{{yes|Windows, Linux, OSX (Sphinx4) / Raspberry-pi, iPhone, Android (pocketsphinx)}}<br /> |[http://cmusphinx.sourceforge.net/ website]<br /> [http://www.cs.cmu.edu/~rsingh/homepage/papers/icassp03-sphinx4_2.pdf paper]<br /> [https://www.cs.cmu.edu/~awb/papers/ICASSP2006/0100185.pdf paper]<br /> [http://sourceforge.net/p/cmusphinx/mailman/ mail-list]<br /> [http://sourceforge.net/p/cmusphinx/discussion/ forum]<br /> [https://github.com/cmusphinx github]<br /> |<br /> |Java (Sphinx4), C (pocketsphinx)<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |CMN, Mel-Spectrum subtraction<br /> |GMM, Streams<br /> |MLLR, MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch<br /> |{{yes|Yes}}<br /> |[http://www.keithv.com/software/sphinx/ AURORA4 (WSJ0)]<br /> |<br /> |-<br /> !HTK<br /> |1993-2009 (3.4.1)<br /> |{{yes|Yes}}<br /> |{{no|[http://htk.eng.cam.ac.uk/docs/license.shtml proprietary]}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://htk.eng.cam.ac.uk/download.shtml website]<br /> [http://htk.eng.cam.ac.uk/docs/docs.shtml book]<br /> [http://htk.eng.cam.ac.uk/mailing/subscribe_mail.shtml mail-list]<br /> |[http://htk.eng.cam.ac.uk/extensions/index.shtml official], [http://htk.eng.cam.ac.uk/develop/atk.shtml ATK], [https://github.com/ramon-astudillo/custom_fe FE uncertainty decoding]<br /> |C<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMN<br /> |GMM (Full Cov.), Tied-Mix, Streams<br /> |HLDA, MLLR (w/ reg. trees), CMLR (w/ adaptive training), MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch, MMI, MPE, MWE<br /> |{{yes|Yes}}<br /> |[http://catalog.elra.info/product_info.php?cPath=37_40&amp;products_id=693 AURORA2 (purch.)] [http://catalog.elra.info/index.php?cPath=37_40 AURORA3 (purch.)], [http://www.keithv.com/software/htk/ AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/projects/chime/PCC/data/pcchome.tar.gz CHIME-1], [ftp://ftp.dcs.shef.ac.uk/share/spandh/chime_challenge/grid/eval_tools_grid.tgz CHIME-2-I], [http://spandh.dcs.shef.ac.uk/chime_challenge/chime2_task2.html CHIME-2-II],[http://reverb2014.dereverberation.com/tools/REVERB_TOOLS_FOR_ASR_ver2.0.tgz REVERB]<br /> |ETSI-AFE-AURORA2 [http://aurora.hsnr.de/download/Aurora2_afe_v1_1.pdf paper] (see AURORA2 purch.)<br /> |-<br /> !Kaldi<br /> |2009-* (continous updates)<br /> |{{yes|Yes}}<br /> |{{Yes|Apache 2.0}}<br /> |{{some|Windows (not mantained as of 2014), Linux, OSX}}<br /> |[http://kaldi.sourceforge.net/about.html website]<br /> [http://homepages.inf.ed.ac.uk/aghoshal/pubs/asru11-kaldi.pdf paper]<br /> [http://sourceforge.net/p/kaldi/mailman/kaldi-users/ mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://kaldi.sourceforge.net/install.html SVN]<br /> |<br /> |C++<br /> |{{yes|BLAS, LAPACK, GPU (for DNNs)}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMVN<br /> |GMM (Full Cov.), SGMM, DNN<br /> |HLDA, STC, MLLT, MLLR, CMLLR (w/ reg. trees), Exponential transform<br /> |aligment, N-best, lattice rescoring (uses OpenFST)<br /> |Baum-Welch, MMI (boosted), MC, feature-based<br /> |{{yes|Yes}}<br /> |[http://kaldi.sourceforge.net/data_prep.html AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/chime_challenge/WSJ0public/CHiME2012-WSJ0-Kaldi_0.03.tar.gz CHIME-2]<br /> |Weniger2014-REVERB [http://reverb2014.dereverberation.com/workshop/reverb2014-papers/1569884459.pdf Paper] [http://www.mmk.ei.tum.de/~wen/REVERB_2014/kaldi_baseline.tar.gz Code]<br /> |-<br /> !Spraak<br /> |2008-* (1.1.374)<br /> |{{yes|Yes}}<br /> |{{no|[http://www.spraak.org/obtaining-spraak/license proprietary]}}<br /> |{{some|Windows ([http://www.spraak.org/obtaining-spraak/system-requirements limited]), Linux, OSX}}<br /> |[http://www.spraak.org/ website]<br /> [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/wambacq/interspeech08/is2008_spraak_v3.pdf paper]<br /> [http://www.spraak.org/mailing-lists mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://www.spraak.org/documentation/doxygen/doc/html/spr__svn.html SVN]<br /> |Missing Data Techniques (MDT)<br /> |C, Python<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |Flexible preprocessing script language -- examples for MFCC, PLP<br /> |VTLN,CMN, [http://www.spraak.org/documentation/doxygen/doc/html/spr__tut__mida.html MIDA], [http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__intro.html MDT Techniques], Parametric HistEq [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/ICASSP2010/zhang.pdf&amp;auto&amp;xz:icassp10], Noise normalization [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |GMM (Tied-Mix), Exemplar based [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/dtw/paper_dtw.pdf&amp;auto&amp;kd:icassp11a], NN, CRF, ... (flexible using the preprocessing script) [http://lib.ugent.be/catalog/pug01:4382368]<br /> |CMLLR, eigenvoices, GMM-weight based (NMF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/SPEECOM2013/manuscript_zhang.pdf&amp;auto&amp;xzhang13] -- (all have Matlab dependencies); MAP<br /> |aligment, lattice rescoring, SCRF rescoring (using SCARF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/scarf/paper_scarf.pdf&amp;auto&amp;kd:icassp11b], phone lattice rescoring [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/duchato/eurospeech09/flavor.pdf&amp;auto&amp;jd:intersp09]<br /> |Viterbi<br /> |{{yes|Yes}}<br /> |[http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__example.html AURORA4], [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |-<br /> !Julius<br /> |1997-* (4.3.1)<br /> |{{yes|Yes}}<br /> |{{no|[http://julius.sourceforge.jp/LICENSE.txt propietary]}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://julius.sourceforge.jp/en_index.php website]<br /> [http://jaist.dl.sourceforge.jp/julius/47534/Juliusbook-4.1.5.pdf book]<br /> [http://julius.sourceforge.jp/juliusbook/en/ book online]<br /> [mailto:julius-info@lists.sourceforge.jp mail-list]<br /> [http://julius.sourceforge.jp/forum/ forum]<br /> [http://julius.sourceforge.jp/en_index.php?q=index-en.html#get_by_cvs CVS]<br /> |[http://prdownloads.sourceforge.jp/julius/23332/slf2dfa-1.0.tar.gz htk2Julius grammar]<br /> [http://sourceforge.jp/projects/julius/downloads/32570/julius4-segmentation-kit-v1.0.tar.gz phoneme seg.]<br /> |C<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> | MFCC<br /> |VTLN,CMVN<br /> |GMM (Tied-Mix)<br /> |<br /> |aligment, two-pass decoder<br /> |Baum-Welch<br /> |{{yes|Yes (low latency)}}<br /> |<br /> |<br /> |-<br /> !RWTH<br /> |2001-* (0.6.1)<br /> |{{yes|Yes}}<br /> |{{some|[http://www-i6.informatik.rwth-aachen.de/rwth-asr/rwth-asr-license.html non-commercial]}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://www-i6.informatik.rwth-aachen.de/rwth-asr/ website]<br /> [http://www.isca-speech.org/archive/interspeech_2009/i09_2111.html paper]<br /> [http://www-i6.informatik.rwth-aachen.de/rwth-asr/manual/ wiki]<br /> [http://www-i6.informatik.rwth-aachen.de/rwth-asr/manual/index.php/Special:AWCforum/%3Faction%3Dsc/id3 forum]<br /> |<br /> |C<br /> |{{yes|BLAS, LAPACK, GPU (CUDA), OpenMP}}<br /> |{{yes|Yes}}<br /> | MFCC, PLP, Gammatone, Tandem (MLP)<br /> |VTLN, CMVN, PCA, LDA<br /> |GMM (Tied covariance), DNN<br /> |MLLR, CMLLR, BIC<br /> |aligment, lattice rescoring, system fusion<br /> |Baum-Welch, MPE<br /> |{{yes|Yes}}<br /> |<br /> |}<br /> '''Other Software'''<br /> <br /> == [[Speaker identification and verification]] ==<br /> <br /> == [[Speech enhancement and separation]] ==<br /> <br /> == [[Other applications]] ==<br /> <br /> == Contribute software ==<br /> To contribute new software, please<br /> * [[Special:UserLogin|create an account]] and login<br /> * go to the wiki page above corresponding to your application; if it does not exist yet, you may create it<br /> * click on the &quot;Edit&quot; link at the top of the page and add a new section for your software (software is ordered by year of the latest version)<br /> * click on the &quot;Save page&quot; link at the bottom of the page to save your modifications<br /> <br /> Please make sure to provide the following information:<br /> * name of the software and year of the latest version<br /> * authors, institution, contact information<br /> * link to the software, ideally including a short demo, and to the external libraries needed<br /> * short description (functionalities, inputs and outputs, programming language, operating system, license, etc) and link to a paper/report describing the software, if any<br /> * whether running on well-known baselines (Aurora-2, Aurora-4, Switchboard, CHiME, etc) is included or requires wrapping by the user<br /> <br /> In order to save storage space, please do not upload the software on this wiki, but link it as much as possible from a public repository (e.g., bitbucket, github, sourceforge) or from a stable URL on the website of your institution. If this is not possible, please contact the [[Main_Page#Working group contacts|resources sharing working group]].</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=List_of_SIG_members&diff=347 List of SIG members 2014-12-03T09:51:52Z <p>Ramon.astudillo: </p> <hr /> <div>111 members as of October 24, 2014:<br /> * Søren Holdt Jensen (Univ. Aalborg)<br /> * Børge Lindberg (Univ. Aalborg)<br /> * Zhen-Hua Tan (Univ. Aalborg)<br /> * Kalle Palomäki (Aalto Univ.)<br /> * Steve Beet (Aculab)<br /> * Gautham Mysore (Adobe)<br /> * Björn Hoffmeister (Amazon)<br /> * Dimitris Dimitriadis (AT&amp;T Research)<br /> * Nassos Katsamanis (TU Athens)<br /> * David Pearce (Audience)<br /> * Wei Xue (Institution of Automation, Chinese Academy of Sciences)<br /> * Keith Ponting (Avaya)<br /> * Daniele Giacobello (Beats by Dr. Dre)<br /> * Armin Sehr (Beuth HS)<br /> * Morgan Nelson (Univ. California, Berkeley)<br /> * Pranav Jawale (IIT Bombay)<br /> * Rajbabu Velmurugan (IIT Bombay)<br /> * Mark Gales (Univ. Cambrigde)<br /> * Angel Mario Castro Martínez (Univ. Carl von Ossietzky)<br /> * Bhiksha Raj (Carnegie-Mellon Univ.)<br /> * Rita Singh (Carnegie-Mellon Univ.)<br /> * Richard Stern (Carnegie-Mellon Univ.)<br /> * Climent Nadeu (UP Catalonia)<br /> * Nestor Becerra Yoma (Univ. Chile)<br /> * Fabio Tesser (CNR)<br /> * Jahangir Alam (CRIM)<br /> * Gerhard Taljaard (CSIR)<br /> * John L. Hansen (UT Dallas)<br /> * Taufiq Hasan (UT Dallas)<br /> * Alexey Tarasov (Dublin IT)<br /> * Emanuël Habets (Univ. Erlangen-Nürnberg)<br /> * Walter Kellermann (Univ. Erlangen-Nürnberg)<br /> * Roland Maas (Univ. Erlangen-Nürnberg)<br /> * Georg Stemmer (Univ. Erlangen-Nürnberg)<br /> * Alessio Brutti (FBK)<br /> * Cristina Guerrero Flores (FBK)<br /> * Waheed Mohamed (FBK)<br /> * Brigitte Richardson (Ford Motors)<br /> * Thad Hughes (Google)<br /> * Kevin Wilson (Google)<br /> * José Andrés Gonzalez (Univ. Granada)<br /> * Antonio M. Peinado (Univ. Granada)<br /> * Pejman Mowlaee (Univ. Graz)<br /> * Kwang Myung Jeon (Gwangju IST)<br /> * Martin Heckmann (Honda)<br /> * Afsaneh Asaei (Idiap)<br /> * Ramya Rasipuram (Idiap)<br /> * Alberto Abad (INESC-Id)<br /> * Ramon F. Astudillo (INESC-Id)<br /> * Arseniy Gorin (Inria)<br /> * Emmanuel Vincent (Inria)<br /> * Dung Tran (Inria)<br /> * Yann Salaün (Inria)<br /> * João Felipe Santos (INRS-EMT)<br /> * Rohit Gowtham (Interactive Intelligence)<br /> * Hynek Hermansky (Johns-Hopkins Univ.)<br /> * Rajesh Hegde (IIT Kanpur)<br /> * Lalan Kumar (IIT Kanpur)<br /> * Karan Nathwani (IIT Kanpur)<br /> * Jort Gemmeke (KU Leuven)<br /> * Hugo van Hamme (KU Leuven)<br /> * Jiri Malek (Univ. Liberec)<br /> * Cong-Thanh Do (LIMSI)<br /> * Rajiv Reddy (LumenVox)<br /> * Alfred Mertins (Univ. Lübeck)<br /> * Florian Müller (Univ. Lübeck)<br /> * Ke Chen (Univ. Manchester)<br /> * Richard Rose (McGill Univ.)<br /> * John Hershey (MERL)<br /> * Jonathan Le Roux (MERL)<br /> * Shinji Watanabe (MERL)<br /> * Mike Seltzer (Microsoft)<br /> * Joseph P. Campbell (MIT)<br /> * Jim Glass (MIT)<br /> * Najim Dehak (MIT)<br /> * Björn Schuller (TU Munich)<br /> * Felix Weninger (TU Munich)<br /> * Lou Boves (Univ. Nijmegen)<br /> * Mario Ganzeboom (Univ. Nijmegen)<br /> * Bert Cranen (Univ. Nijmegen)<br /> * Louis ten Bosch (Univ. Nijmegen)<br /> * David van Leeuwen (Univ. Nijmegen)<br /> * Shuguang Wang (Northwestern Polytechnical Univ., Xi'an)<br /> * Marc Delcroix (NTT)<br /> * Miquel Espi (NTT)<br /> * Keisuke Kinoshita (NTT)<br /> * Cyril Guillaumé (NXP Software)<br /> * Wouter Tirry (NXP Software)<br /> * Arun Narayanan (Ohio State Univ.)<br /> * DeLiang Wang (Ohio State Univ.)<br /> * Simon Bozonnet (OnMobile)<br /> * Reinhold Häb-Umbach (Univ. Paderborn)<br /> * Ananya Misra (Princeton Univ.)<br /> * Ricard Marxer (Univ. of Sheffield)<br /> * Ahmed Hussen Abdelaziz (Ruhr Univ., Bochum)<br /> * Dorothea Kolossa (Ruhr Univ., Bochum)<br /> * Rahil Mahdiyan (Saarland Univ.)<br /> * Hakan Erdogan (Sabanci Univ.)<br /> * Emad Grais (Sabanci Univ.)<br /> * Raghavendra Bilgi (Samsung)<br /> * David Trainor (Sentireal)<br /> * Kenichi Kumatani (Spansion)<br /> * Jordan Cohen (Spelamode)<br /> * Leslie Smith (Univ. Stirling)<br /> * Gérard Chollet (Télécom ParisTech)<br /> * Otman Chakkor (Univ. Tetuan)<br /> * Ngoc Q.K. Duong (Technicolor)<br /> * Zhijian Ou (Tsinghua Univ.)<br /> * Antti Hurmalainen (TU Tampere)<br /> * Tuomas Virtanen (TU Tampere)<br /> * Vinay Modi (Voice Pitara Technologies)<br /> * Jidong Tao (Vlingo)<br /> * Raphaël Blouet</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=Software&diff=344 Software 2014-11-27T18:36:59Z <p>Ramon.astudillo: /* Automatic speech recognition */</p> <hr /> <div>This page provides software grouped by application. <br /> <br /> == [[Automatic speech recognition]] ==<br /> <br /> {| class=&quot;wikitable sortable&quot; style=&quot;font-size:72%; border:gray solid 1px; text-align:center; width:auto; table-layout:fixed;&quot;<br /> |-<br /> !style=&quot;width: 40px&quot; rowspan=&quot;2&quot; class=&quot;unsortable&quot;| ASR engines<br /> !colspan=&quot;6&quot; |General attributes<br /> !colspan=&quot;2&quot; |Programming<br /> !colspan=&quot;8&quot; |Implemented ASR techniques<br /> !colspan=&quot;2&quot; |Reproducible research<br /> |-<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | release / update<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | actively developed<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | licence<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | platforms<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | links<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | extensions<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | language<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | hardware optimization<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | VAD<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic features<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | feature normalization / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic models<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | model adaptation / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | decoding techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | training techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | online ASR<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | robust ASR training recipes<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | reproducible results<br /> |-<br /> !CMU Sphinx<br /> |1986-* (Sphinx 4.1.0, pocketsphinx 0.8)<br /> |{{yes|Yes}}<br /> |{{yes|[https://raw.githubusercontent.com/cmusphinx/sphinx4/master/license.terms BSD-like]}}<br /> |{{yes|Windows, Linux, OSX (Sphinx4) / Raspberry-pi, iPhone, Android (pocketsphinx)}}<br /> |[http://cmusphinx.sourceforge.net/ website]<br /> [http://www.cs.cmu.edu/~rsingh/homepage/papers/icassp03-sphinx4_2.pdf paper]<br /> [https://www.cs.cmu.edu/~awb/papers/ICASSP2006/0100185.pdf paper]<br /> [http://sourceforge.net/p/cmusphinx/mailman/ mail-list]<br /> [http://sourceforge.net/p/cmusphinx/discussion/ forum]<br /> [https://github.com/cmusphinx github]<br /> |<br /> |Java (Sphinx4), C (pocketsphinx)<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |CMN, Mel-Spectrum subtraction<br /> |GMM, Streams<br /> |MLLR, MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch<br /> |{{yes|Yes}}<br /> |[http://www.keithv.com/software/sphinx/ AURORA4 (WSJ0)]<br /> |<br /> |-<br /> !HTK<br /> |1993-2009 (3.4.1)<br /> |{{yes|Yes}}<br /> |{{no|[http://htk.eng.cam.ac.uk/docs/license.shtml proprietary]}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://htk.eng.cam.ac.uk/download.shtml website]<br /> [http://htk.eng.cam.ac.uk/docs/docs.shtml book]<br /> [http://htk.eng.cam.ac.uk/mailing/subscribe_mail.shtml mail-list]<br /> |[http://htk.eng.cam.ac.uk/extensions/index.shtml official], [http://htk.eng.cam.ac.uk/develop/atk.shtml ATK], [https://github.com/ramon-astudillo/custom_fe FE uncertainty decoding]<br /> |C<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMN<br /> |GMM (Full Cov.), Tied-Mix, Streams<br /> |HLDA, MLLR (w/ reg. trees), CMLR (w/ adaptive training), MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch, MMI, MPE, MWE<br /> |{{yes|Yes}}<br /> |[http://catalog.elra.info/product_info.php?cPath=37_40&amp;products_id=693 AURORA2 (purch.)] [http://catalog.elra.info/index.php?cPath=37_40 AURORA3 (purch.)], [http://www.keithv.com/software/htk/ AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/projects/chime/PCC/data/pcchome.tar.gz CHIME-1], [ftp://ftp.dcs.shef.ac.uk/share/spandh/chime_challenge/grid/eval_tools_grid.tgz CHIME-2-I], [http://spandh.dcs.shef.ac.uk/chime_challenge/chime2_task2.html CHIME-2-II],[http://reverb2014.dereverberation.com/tools/REVERB_TOOLS_FOR_ASR_ver2.0.tgz REVERB]<br /> |ETSI-AFE-AURORA2 [http://aurora.hsnr.de/download/Aurora2_afe_v1_1.pdf paper] (see AURORA2 purch.)<br /> |-<br /> !Kaldi<br /> |2009-* (continous updates)<br /> |{{yes|Yes}}<br /> |{{Yes|Apache 2.0}}<br /> |{{some|Windows (not mantained as of 2014), Linux, OSX}}<br /> |[http://kaldi.sourceforge.net/about.html website]<br /> [http://homepages.inf.ed.ac.uk/aghoshal/pubs/asru11-kaldi.pdf paper]<br /> [http://sourceforge.net/p/kaldi/mailman/kaldi-users/ mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://kaldi.sourceforge.net/install.html SVN]<br /> |<br /> |C++<br /> |{{yes|BLAS, LAPACK, GPU (for DNNs)}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMVN<br /> |GMM (Full Cov.), SGMM, DNN<br /> |HLDA, STC, MLLT, MLLR, CMLLR (w/ reg. trees), Exponential transform<br /> |aligment, N-best, lattice rescoring (uses OpenFST)<br /> |Baum-Welch, MMI (boosted), MC, feature-based<br /> |{{yes|Yes}}<br /> |[http://kaldi.sourceforge.net/data_prep.html AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/chime_challenge/WSJ0public/CHiME2012-WSJ0-Kaldi_0.03.tar.gz CHIME-2]<br /> |Weniger2014-REVERB [http://reverb2014.dereverberation.com/workshop/reverb2014-papers/1569884459.pdf Paper] [http://www.mmk.ei.tum.de/~wen/REVERB_2014/kaldi_baseline.tar.gz Code]<br /> |-<br /> !Spraak<br /> |2008-* (1.1.374)<br /> |{{yes|Yes}}<br /> |{{no|[http://www.spraak.org/obtaining-spraak/license proprietary]}}<br /> |{{some|Windows ([http://www.spraak.org/obtaining-spraak/system-requirements limited]), Linux, OSX}}<br /> |[http://www.spraak.org/ website]<br /> [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/wambacq/interspeech08/is2008_spraak_v3.pdf paper]<br /> [http://www.spraak.org/mailing-lists mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://www.spraak.org/documentation/doxygen/doc/html/spr__svn.html SVN]<br /> |Missing Data Techniques (MDT)<br /> |C, Python<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |Flexible preprocessing script language -- examples for MFCC, PLP<br /> |VTLN,CMN, [http://www.spraak.org/documentation/doxygen/doc/html/spr__tut__mida.html MIDA], [http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__intro.html MDT Techniques], Parametric HistEq [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/ICASSP2010/zhang.pdf&amp;auto&amp;xz:icassp10], Noise normalization [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |GMM (Tied-Mix), Exemplar based [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/dtw/paper_dtw.pdf&amp;auto&amp;kd:icassp11a], NN, CRF, ... (flexible using the preprocessing script) [http://lib.ugent.be/catalog/pug01:4382368]<br /> |CMLLR, eigenvoices, GMM-weight based (NMF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/SPEECOM2013/manuscript_zhang.pdf&amp;auto&amp;xzhang13] -- (all have Matlab dependencies); MAP<br /> |aligment, lattice rescoring, SCRF rescoring (using SCARF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/scarf/paper_scarf.pdf&amp;auto&amp;kd:icassp11b], phone lattice rescoring [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/duchato/eurospeech09/flavor.pdf&amp;auto&amp;jd:intersp09]<br /> |Viterbi<br /> |{{yes|Yes}}<br /> |[http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__example.html AURORA4], [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |-<br /> !Julius<br /> |1997-* (4.3.1)<br /> |{{yes|Yes}}<br /> |{{no|[http://julius.sourceforge.jp/LICENSE.txt propietary]}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://julius.sourceforge.jp/en_index.php website]<br /> [http://jaist.dl.sourceforge.jp/julius/47534/Juliusbook-4.1.5.pdf book]<br /> [http://julius.sourceforge.jp/juliusbook/en/ book online]<br /> [mailto:julius-info@lists.sourceforge.jp mail-list]<br /> [http://julius.sourceforge.jp/forum/ forum]<br /> [http://julius.sourceforge.jp/en_index.php?q=index-en.html#get_by_cvs CVS]<br /> |[http://prdownloads.sourceforge.jp/julius/23332/slf2dfa-1.0.tar.gz htk2Julius grammar]<br /> [http://sourceforge.jp/projects/julius/downloads/32570/julius4-segmentation-kit-v1.0.tar.gz phoneme seg.]<br /> |C<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> | MFCC<br /> |VTLN,CMVN<br /> |GMM (Tied-Mix)<br /> |<br /> |aligment, two-pass decoder<br /> |Baum-Welch<br /> |{{yes|Yes (low latency)}}<br /> |<br /> |<br /> |-<br /> !RWTH<br /> |2001-* (0.6.1)<br /> |{{yes|Yes}}<br /> |{{some|[http://www-i6.informatik.rwth-aachen.de/rwth-asr/rwth-asr-license.html non-commercial]}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://www-i6.informatik.rwth-aachen.de/rwth-asr/ website]<br /> [http://www.isca-speech.org/archive/interspeech_2009/i09_2111.html paper]<br /> [http://www-i6.informatik.rwth-aachen.de/rwth-asr/manual/ wiki]<br /> [http://www-i6.informatik.rwth-aachen.de/rwth-asr/manual/index.php/Special:AWCforum/%3Faction%3Dsc/id3 forum]<br /> |<br /> |C<br /> |{{yes|BLAS, LAPACK, GPU (CUDA), OpenMP}}<br /> |{{yes|Yes}}<br /> | MFCC, PLP, Gammatone, Tandem (MLP)<br /> |VTLN, CMVN, PCA, LDA<br /> |GMM (Tied covariance), DNN<br /> |MLLR, CMLLR, BIC<br /> |aligment, lattice rescoring, system fusion<br /> |Baum-Welch, MPE<br /> |{{yes|Yes}}<br /> |<br /> |}<br /> <br /> == [[Speaker identification and verification]] ==<br /> <br /> == [[Speech enhancement and separation]] ==<br /> <br /> == [[Other applications]] ==<br /> <br /> == Contribute software ==<br /> To contribute new software, please<br /> * [[Special:UserLogin|create an account]] and login<br /> * go to the wiki page above corresponding to your application; if it does not exist yet, you may create it<br /> * click on the &quot;Edit&quot; link at the top of the page and add a new section for your software (software is ordered by year of the latest version)<br /> * click on the &quot;Save page&quot; link at the bottom of the page to save your modifications<br /> <br /> Please make sure to provide the following information:<br /> * name of the software and year of the latest version<br /> * authors, institution, contact information<br /> * link to the software, ideally including a short demo, and to the external libraries needed<br /> * short description (functionalities, inputs and outputs, programming language, operating system, license, etc) and link to a paper/report describing the software, if any<br /> * whether running on well-known baselines (Aurora-2, Aurora-4, Switchboard, CHiME, etc) is included or requires wrapping by the user<br /> <br /> In order to save storage space, please do not upload the software on this wiki, but link it as much as possible from a public repository (e.g., bitbucket, github, sourceforge) or from a stable URL on the website of your institution. If this is not possible, please contact the [[Main_Page#Working group contacts|resources sharing working group]].</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=Software&diff=328 Software 2014-11-26T09:59:26Z <p>Ramon.astudillo: /* Automatic speech recognition */</p> <hr /> <div>This page provides software grouped by application. <br /> <br /> == [[Automatic speech recognition]] ==<br /> <br /> {| class=&quot;wikitable sortable&quot; style=&quot;font-size:72%; border:gray solid 1px; text-align:center; width:auto; table-layout:fixed;&quot;<br /> |-<br /> !style=&quot;width: 40px&quot; rowspan=&quot;2&quot; class=&quot;unsortable&quot;| ASR engines<br /> !colspan=&quot;6&quot; |General attributes<br /> !colspan=&quot;2&quot; |Programming<br /> !colspan=&quot;8&quot; |Implemented ASR techniques<br /> !colspan=&quot;2&quot; |Reproducible research<br /> |-<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | release / update<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | actively developed<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | licence<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | platforms<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | links <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | extensions<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | language<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | hardware optimization<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | VAD<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic features<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | feature normalization / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic models<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | model adaptation / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | decoding techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | training techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | online ASR<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | robust ASR training recipes <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | reproducible results<br /> |-<br /> !CMU Sphinx<br /> |1986-* (Sphinx 4.1.0, pocketsphinx 0.8) <br /> |{{yes|Yes}}<br /> |{{yes|[https://raw.githubusercontent.com/cmusphinx/sphinx4/master/license.terms BSD-like]}}<br /> |{{yes|Windows, Linux, OSX (Sphinx4) / Raspberry-pi, iPhone, Android (pocketsphinx)}}<br /> |[http://cmusphinx.sourceforge.net/ website] <br /> [http://www.cs.cmu.edu/~rsingh/homepage/papers/icassp03-sphinx4_2.pdf paper]<br /> [https://www.cs.cmu.edu/~awb/papers/ICASSP2006/0100185.pdf paper]<br /> [http://sourceforge.net/p/cmusphinx/mailman/ mail-list]<br /> [http://sourceforge.net/p/cmusphinx/discussion/ forum]<br /> [https://github.com/cmusphinx github]<br /> |<br /> |Java (Sphinx4), C (pocketsphinx)<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |CMN, Mel-Spectrum subtraction<br /> |GMM, Streams<br /> |MLLR, MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch<br /> |{{yes|Yes}}<br /> |[http://www.keithv.com/software/sphinx/ AURORA4 (WSJ0)]<br /> |<br /> |-<br /> !HTK<br /> |1993-2009 (3.4.1) <br /> |{{yes|Yes}}<br /> |{{no|[http://htk.eng.cam.ac.uk/docs/license.shtml proprietary]}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://htk.eng.cam.ac.uk/download.shtml website]<br /> [http://htk.eng.cam.ac.uk/docs/docs.shtml book]<br /> [http://htk.eng.cam.ac.uk/mailing/subscribe_mail.shtml mail-list]<br /> |[http://htk.eng.cam.ac.uk/extensions/index.shtml official], [http://htk.eng.cam.ac.uk/develop/atk.shtml ATK], [https://github.com/ramon-astudillo/custom_fe FE uncertainty decoding]<br /> |C<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMN<br /> |GMM (Full Cov.), Tied-Mix, Streams<br /> |HLDA, MLLR (w/ reg. trees), CMLR (w/ adaptive training), MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch, MMI, MPE, MWE<br /> |{{yes|Yes}}<br /> |[http://catalog.elra.info/product_info.php?cPath=37_40&amp;products_id=693 AURORA2 (purch.)] [http://catalog.elra.info/index.php?cPath=37_40 AURORA3 (purch.)], [http://www.keithv.com/software/htk/ AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/projects/chime/PCC/data/pcchome.tar.gz CHIME-1], [ftp://ftp.dcs.shef.ac.uk/share/spandh/chime_challenge/grid/eval_tools_grid.tgz CHIME-2-I], [http://spandh.dcs.shef.ac.uk/chime_challenge/chime2_task2.html CHIME-2-II],[http://reverb2014.dereverberation.com/tools/REVERB_TOOLS_FOR_ASR_ver2.0.tgz REVERB]<br /> |ETSI-AFE-AURORA2 [http://aurora.hsnr.de/download/Aurora2_afe_v1_1.pdf paper] (see AURORA2 purch.)<br /> |-<br /> !Kaldi<br /> |2009-* (continous updates)<br /> |{{yes|Yes}}<br /> |{{Yes|Apache 2.0}}<br /> |{{some|Windows (not mantained as of 2014), Linux, OSX}}<br /> |[http://kaldi.sourceforge.net/about.html website]<br /> [http://homepages.inf.ed.ac.uk/aghoshal/pubs/asru11-kaldi.pdf paper]<br /> [http://sourceforge.net/p/kaldi/mailman/kaldi-users/ mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://kaldi.sourceforge.net/install.html SVN]<br /> |<br /> |C++<br /> |{{yes|BLAS, LAPACK, GPU (for DNNs)}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMVN<br /> |GMM (Full Cov.), SGMM, DNN<br /> |HLDA, STC, MLLT, MLLR, CMLLR (w/ reg. trees), Exponential transform<br /> |aligment, N-best, lattice rescoring (using OpenFST)<br /> |Baum-Welch, MMI (boosted), MC, feature-based<br /> |{{yes|Yes}}<br /> |[http://kaldi.sourceforge.net/data_prep.html AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/chime_challenge/WSJ0public/CHiME2012-WSJ0-Kaldi_0.03.tar.gz CHIME-2]<br /> |Weniger2014-REVERB [http://reverb2014.dereverberation.com/workshop/reverb2014-papers/1569884459.pdf Paper] [http://www.mmk.ei.tum.de/~wen/REVERB_2014/kaldi_baseline.tar.gz Code]<br /> |-<br /> !Spraak<br /> |2008-* (1.1.374)<br /> |{{yes|Yes}}<br /> |{{no|[http://www.spraak.org/obtaining-spraak/license proprietary2]}}<br /> |{{some|Windows ([http://www.spraak.org/obtaining-spraak/system-requirements limited]), Linux, OSX}}<br /> |[http://www.spraak.org/ website]<br /> [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/wambacq/interspeech08/is2008_spraak_v3.pdf paper]<br /> [http://www.spraak.org/mailing-lists mail-list]<br /> [http://www.spraak.org/mailing-lists mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://www.spraak.org/documentation/doxygen/doc/html/spr__svn.html SVN]<br /> |Missing Data Techniques (MDT)<br /> |C, Python<br /> |{{no|No}} <br /> |{{yes|Yes}}<br /> |Flexible preprocessing script language -- examples for MFCC, PLP <br /> |VTLN,CMN, [http://www.spraak.org/documentation/doxygen/doc/html/spr__tut__mida.html MIDA], [http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__intro.html MDT Techniques], Parametric HistEq [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/ICASSP2010/zhang.pdf&amp;auto&amp;xz:icassp10], Noise normalization [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |GMM (Tied-Mix), Exemplar based [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/dtw/paper_dtw.pdf&amp;auto&amp;kd:icassp11a], NN, CRF, ... (flexible using the preprocessing script) [http://lib.ugent.be/catalog/pug01:4382368]<br /> |CMLLR, eigenvoices, GMM-weight based (NMF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/SPEECOM2013/manuscript_zhang.pdf&amp;auto&amp;xzhang13] -- (all have Matlab dependencies); MAP<br /> |aligment, lattice rescoring, SCRF rescoring (using SCARF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/scarf/paper_scarf.pdf&amp;auto&amp;kd:icassp11b], phone lattice rescoring [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/duchato/eurospeech09/flavor.pdf&amp;auto&amp;jd:intersp09]<br /> |Viterbi<br /> |{{yes|Yes}<br /> |[http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__example.html AURORA4], [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |<br /> |}<br /> <br /> == [[Speaker identification and verification]] ==<br /> <br /> == [[Speech enhancement and separation]] ==<br /> <br /> == [[Other applications]] ==<br /> <br /> == Contribute software ==<br /> To contribute new software, please<br /> * [[Special:UserLogin|create an account]] and login<br /> * go to the wiki page above corresponding to your application; if it does not exist yet, you may create it<br /> * click on the &quot;Edit&quot; link at the top of the page and add a new section for your software (software is ordered by year of the latest version)<br /> * click on the &quot;Save page&quot; link at the bottom of the page to save your modifications<br /> <br /> Please make sure to provide the following information:<br /> * name of the software and year of the latest version<br /> * authors, institution, contact information<br /> * link to the software, ideally including a short demo, and to the external libraries needed<br /> * short description (functionalities, inputs and outputs, programming language, operating system, license, etc) and link to a paper/report describing the software, if any<br /> * whether running on well-known baselines (Aurora-2, Aurora-4, Switchboard, CHiME, etc) is included or requires wrapping by the user<br /> <br /> In order to save storage space, please do not upload the software on this wiki, but link it as much as possible from a public repository (e.g., bitbucket, github, sourceforge) or from a stable URL on the website of your institution. If this is not possible, please contact the [[Main_Page#Working group contacts|resources sharing working group]].</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=Software&diff=327 Software 2014-11-26T09:59:14Z <p>Ramon.astudillo: /* Automatic speech recognition */</p> <hr /> <div>This page provides software grouped by application. <br /> <br /> == [[Automatic speech recognition]] ==<br /> <br /> {| class=&quot;wikitable sortable&quot; style=&quot;font-size:72%; border:gray solid 1px; text-align:center; width:auto; table-layout:fixed;&quot;<br /> |-<br /> !style=&quot;width: 40px&quot; rowspan=&quot;2&quot; class=&quot;unsortable&quot;| ASR engines<br /> !colspan=&quot;6&quot; |General attributes<br /> !colspan=&quot;2&quot; |Programming<br /> !colspan=&quot;8&quot; |Implemented ASR techniques<br /> !colspan=&quot;2&quot; |Reproducible research<br /> |-<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | release / update<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | actively developed<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | licence<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | platforms<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | links <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | extensions<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | language<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | hardware optimization<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | VAD<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic features<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | feature normalization / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic models<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | model adaptation / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | decoding techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | training techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | online ASR<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | robust ASR training recipes <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | reproducible results<br /> |-<br /> !CMU Sphinx<br /> |1986-* (Sphinx 4.1.0, pocketsphinx 0.8) <br /> |{{yes|Yes}}<br /> |{{yes|[https://raw.githubusercontent.com/cmusphinx/sphinx4/master/license.terms BSD-like]}}<br /> |{{yes|Windows, Linux, OSX (Sphinx4) / Raspberry-pi, iPhone, Android (pocketsphinx)}}<br /> |[http://cmusphinx.sourceforge.net/ website] <br /> [http://www.cs.cmu.edu/~rsingh/homepage/papers/icassp03-sphinx4_2.pdf paper]<br /> [https://www.cs.cmu.edu/~awb/papers/ICASSP2006/0100185.pdf paper]<br /> [http://sourceforge.net/p/cmusphinx/mailman/ mail-list]<br /> [http://sourceforge.net/p/cmusphinx/discussion/ forum]<br /> [https://github.com/cmusphinx github]<br /> |<br /> |Java (Sphinx4), C (pocketsphinx)<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |CMN, Mel-Spectrum subtraction<br /> |GMM, Streams<br /> |MLLR, MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch<br /> |{{yes|Yes}}<br /> |[http://www.keithv.com/software/sphinx/ AURORA4 (WSJ0)]<br /> |<br /> |-<br /> !HTK<br /> |1993-2009 (3.4.1) <br /> |{{yes|Yes}}<br /> |{{no|[http://htk.eng.cam.ac.uk/docs/license.shtml proprietary]}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://htk.eng.cam.ac.uk/download.shtml website]<br /> [http://htk.eng.cam.ac.uk/docs/docs.shtml book]<br /> [http://htk.eng.cam.ac.uk/docs/docs.shtml book]<br /> [http://htk.eng.cam.ac.uk/mailing/subscribe_mail.shtml mail-list]<br /> |[http://htk.eng.cam.ac.uk/extensions/index.shtml official], [http://htk.eng.cam.ac.uk/develop/atk.shtml ATK], [https://github.com/ramon-astudillo/custom_fe FE uncertainty decoding]<br /> |C<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMN<br /> |GMM (Full Cov.), Tied-Mix, Streams<br /> |HLDA, MLLR (w/ reg. trees), CMLR (w/ adaptive training), MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch, MMI, MPE, MWE<br /> |{{yes|Yes}}<br /> |[http://catalog.elra.info/product_info.php?cPath=37_40&amp;products_id=693 AURORA2 (purch.)] [http://catalog.elra.info/index.php?cPath=37_40 AURORA3 (purch.)], [http://www.keithv.com/software/htk/ AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/projects/chime/PCC/data/pcchome.tar.gz CHIME-1], [ftp://ftp.dcs.shef.ac.uk/share/spandh/chime_challenge/grid/eval_tools_grid.tgz CHIME-2-I], [http://spandh.dcs.shef.ac.uk/chime_challenge/chime2_task2.html CHIME-2-II],[http://reverb2014.dereverberation.com/tools/REVERB_TOOLS_FOR_ASR_ver2.0.tgz REVERB]<br /> |ETSI-AFE-AURORA2 [http://aurora.hsnr.de/download/Aurora2_afe_v1_1.pdf paper] (see AURORA2 purch.)<br /> |-<br /> !Kaldi<br /> |2009-* (continous updates)<br /> |{{yes|Yes}}<br /> |{{Yes|Apache 2.0}}<br /> |{{some|Windows (not mantained as of 2014), Linux, OSX}}<br /> |[http://kaldi.sourceforge.net/about.html website]<br /> [http://homepages.inf.ed.ac.uk/aghoshal/pubs/asru11-kaldi.pdf paper]<br /> [http://sourceforge.net/p/kaldi/mailman/kaldi-users/ mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://kaldi.sourceforge.net/install.html SVN]<br /> |<br /> |C++<br /> |{{yes|BLAS, LAPACK, GPU (for DNNs)}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMVN<br /> |GMM (Full Cov.), SGMM, DNN<br /> |HLDA, STC, MLLT, MLLR, CMLLR (w/ reg. trees), Exponential transform<br /> |aligment, N-best, lattice rescoring (using OpenFST)<br /> |Baum-Welch, MMI (boosted), MC, feature-based<br /> |{{yes|Yes}}<br /> |[http://kaldi.sourceforge.net/data_prep.html AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/chime_challenge/WSJ0public/CHiME2012-WSJ0-Kaldi_0.03.tar.gz CHIME-2]<br /> |Weniger2014-REVERB [http://reverb2014.dereverberation.com/workshop/reverb2014-papers/1569884459.pdf Paper] [http://www.mmk.ei.tum.de/~wen/REVERB_2014/kaldi_baseline.tar.gz Code]<br /> |-<br /> !Spraak<br /> |2008-* (1.1.374)<br /> |{{yes|Yes}}<br /> |{{no|[http://www.spraak.org/obtaining-spraak/license proprietary2]}}<br /> |{{some|Windows ([http://www.spraak.org/obtaining-spraak/system-requirements limited]), Linux, OSX}}<br /> |[http://www.spraak.org/ website]<br /> [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/wambacq/interspeech08/is2008_spraak_v3.pdf paper]<br /> [http://www.spraak.org/mailing-lists mail-list]<br /> [http://www.spraak.org/mailing-lists mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://www.spraak.org/documentation/doxygen/doc/html/spr__svn.html SVN]<br /> |Missing Data Techniques (MDT)<br /> |C, Python<br /> |{{no|No}} <br /> |{{yes|Yes}}<br /> |Flexible preprocessing script language -- examples for MFCC, PLP <br /> |VTLN,CMN, [http://www.spraak.org/documentation/doxygen/doc/html/spr__tut__mida.html MIDA], [http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__intro.html MDT Techniques], Parametric HistEq [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/ICASSP2010/zhang.pdf&amp;auto&amp;xz:icassp10], Noise normalization [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |GMM (Tied-Mix), Exemplar based [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/dtw/paper_dtw.pdf&amp;auto&amp;kd:icassp11a], NN, CRF, ... (flexible using the preprocessing script) [http://lib.ugent.be/catalog/pug01:4382368]<br /> |CMLLR, eigenvoices, GMM-weight based (NMF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/SPEECOM2013/manuscript_zhang.pdf&amp;auto&amp;xzhang13] -- (all have Matlab dependencies); MAP<br /> |aligment, lattice rescoring, SCRF rescoring (using SCARF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/scarf/paper_scarf.pdf&amp;auto&amp;kd:icassp11b], phone lattice rescoring [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/duchato/eurospeech09/flavor.pdf&amp;auto&amp;jd:intersp09]<br /> |Viterbi<br /> |{{yes|Yes}<br /> |[http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__example.html AURORA4], [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |<br /> |}<br /> <br /> == [[Speaker identification and verification]] ==<br /> <br /> == [[Speech enhancement and separation]] ==<br /> <br /> == [[Other applications]] ==<br /> <br /> == Contribute software ==<br /> To contribute new software, please<br /> * [[Special:UserLogin|create an account]] and login<br /> * go to the wiki page above corresponding to your application; if it does not exist yet, you may create it<br /> * click on the &quot;Edit&quot; link at the top of the page and add a new section for your software (software is ordered by year of the latest version)<br /> * click on the &quot;Save page&quot; link at the bottom of the page to save your modifications<br /> <br /> Please make sure to provide the following information:<br /> * name of the software and year of the latest version<br /> * authors, institution, contact information<br /> * link to the software, ideally including a short demo, and to the external libraries needed<br /> * short description (functionalities, inputs and outputs, programming language, operating system, license, etc) and link to a paper/report describing the software, if any<br /> * whether running on well-known baselines (Aurora-2, Aurora-4, Switchboard, CHiME, etc) is included or requires wrapping by the user<br /> <br /> In order to save storage space, please do not upload the software on this wiki, but link it as much as possible from a public repository (e.g., bitbucket, github, sourceforge) or from a stable URL on the website of your institution. If this is not possible, please contact the [[Main_Page#Working group contacts|resources sharing working group]].</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=Main_Page&diff=326 Main Page 2014-11-26T09:58:51Z <p>Ramon.astudillo: /* Best practices */</p> <hr /> <div>Welcome to the wiki of ISCA's Special Interest Group (SIG) on Robust Speech Processing. This wiki aims to be a repository of resources for researchers interested in Robust Speech Processing.<br /> <br /> == Contribute ==<br /> All the material is contributed by users. Please contribute and advertise your own resources! This will take you a few minutes only. To do so, just click on one of the links below and follow the guidelines.<br /> <br /> == Join the SIG - Google/LinkedIn groups==<br /> You are encouraged to join the SIG to benefit from our announcements of recent research developments and other matters of interest in Robust Speech Processing. See the [[Constitution of the ISCA Special Interest Group on Robust Speech Processing|SIG constitution]] for more information. Membership is free of charge and ISCA membership is not required.<br /> <br /> '''To join''', just register to our [https://groups.google.com/forum/#!forum/rosp-sig Google group].<br /> <br /> You are also welcome to join our [https://www.linkedin.com/groups/Robust-Speech-Processing-6744473 LinkeIn group]<br /> <br /> == [[Datasets]] ==<br /> This page provides a list of datasets, with corresponding software baselines and evaluation results.<br /> <br /> == [[Software]] ==<br /> This page provides a list of speech processing software and corresponding application recipes for different datasets.<br /> <br /> == [[Best practices]] ==<br /> <br /> == [[SIG events]] ==<br /> <br /> [http://hscma2014.inria.fr/ HSCMA 2014]<br /> <br /> == Other resources ==<br /> [[Constitution of the ISCA Special Interest Group on Robust Speech Processing|RoSP SIG constitution]]<br /> <br /> [[List of SIG members]]<br /> <br /> == Working group contacts ==<br /> Resources sharing working group: [mailto:ramon@astudillo.com Ramon F. Astudillo], [mailto:jgemmeke@amadana.nl Jort F. Gemmeke], [mailto:emmanuel.vincent@inria.fr Emmanuel Vincent].<br /> <br /> Best practices working group: [mailto:ramon@astudillo.com Ramon F. Astudillo], [mailto:john.hansen@utdallas.edu John L. Hansen], [mailto:zt@es.aau.dk Zheng-Hua Tan].</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=Main_Page&diff=325 Main Page 2014-11-26T09:57:36Z <p>Ramon.astudillo: /* Best practices */</p> <hr /> <div>Welcome to the wiki of ISCA's Special Interest Group (SIG) on Robust Speech Processing. This wiki aims to be a repository of resources for researchers interested in Robust Speech Processing.<br /> <br /> == Contribute ==<br /> All the material is contributed by users. Please contribute and advertise your own resources! This will take you a few minutes only. To do so, just click on one of the links below and follow the guidelines.<br /> <br /> == Join the SIG - Google/LinkedIn groups==<br /> You are encouraged to join the SIG to benefit from our announcements of recent research developments and other matters of interest in Robust Speech Processing. See the [[Constitution of the ISCA Special Interest Group on Robust Speech Processing|SIG constitution]] for more information. Membership is free of charge and ISCA membership is not required.<br /> <br /> '''To join''', just register to our [https://groups.google.com/forum/#!forum/rosp-sig Google group].<br /> <br /> You are also welcome to join our [https://www.linkedin.com/groups/Robust-Speech-Processing-6744473 LinkeIn group]<br /> <br /> == [[Datasets]] ==<br /> This page provides a list of datasets, with corresponding software baselines and evaluation results.<br /> <br /> == [[Software]] ==<br /> This page provides a list of speech processing software and corresponding application recipes for different datasets.<br /> <br /> == [[Best practices]] ==<br /> some text<br /> <br /> == [[SIG events]] ==<br /> <br /> [http://hscma2014.inria.fr/ HSCMA 2014]<br /> <br /> == Other resources ==<br /> [[Constitution of the ISCA Special Interest Group on Robust Speech Processing|RoSP SIG constitution]]<br /> <br /> [[List of SIG members]]<br /> <br /> == Working group contacts ==<br /> Resources sharing working group: [mailto:ramon@astudillo.com Ramon F. Astudillo], [mailto:jgemmeke@amadana.nl Jort F. Gemmeke], [mailto:emmanuel.vincent@inria.fr Emmanuel Vincent].<br /> <br /> Best practices working group: [mailto:ramon@astudillo.com Ramon F. Astudillo], [mailto:john.hansen@utdallas.edu John L. Hansen], [mailto:zt@es.aau.dk Zheng-Hua Tan].</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=Software&diff=324 Software 2014-11-26T09:10:27Z <p>Ramon.astudillo: /* Automatic speech recognition */</p> <hr /> <div>This page provides software grouped by application. <br /> <br /> == [[Automatic speech recognition]] ==<br /> <br /> {| class=&quot;wikitable sortable&quot; style=&quot;font-size:72%; border:gray solid 1px; text-align:center; width:auto; table-layout:fixed;&quot;<br /> |-<br /> !style=&quot;width: 40px&quot; rowspan=&quot;2&quot; class=&quot;unsortable&quot;| ASR engines<br /> !colspan=&quot;6&quot; |General attributes<br /> !colspan=&quot;2&quot; |Programming<br /> !colspan=&quot;8&quot; |Implemented ASR techniques<br /> !colspan=&quot;2&quot; |Reproducible research<br /> |-<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | release / update<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | actively developed<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | licence<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | platforms<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | links <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | extensions<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | language<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | hardware optimization<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | VAD<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic features<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | feature normalization / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic models<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | model adaptation / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | decoding techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | training techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | online ASR<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | robust ASR training recipes <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | reproducible results<br /> |-<br /> !CMU Sphinx<br /> |1986-* (Sphinx 4.1.0, pocketsphinx 0.8) <br /> |{{yes|Yes}}<br /> |{{yes|[https://raw.githubusercontent.com/cmusphinx/sphinx4/master/license.terms BSD-like]}}<br /> |{{yes|Windows, Linux, OSX (Sphinx4) / Raspberry-pi, iPhone, Android (pocketsphinx)}}<br /> |[http://cmusphinx.sourceforge.net/ website] <br /> [http://www.cs.cmu.edu/~rsingh/homepage/papers/icassp03-sphinx4_2.pdf paper]<br /> [https://www.cs.cmu.edu/~awb/papers/ICASSP2006/0100185.pdf paper]<br /> [http://sourceforge.net/p/cmusphinx/mailman/ mail-list]<br /> [http://sourceforge.net/p/cmusphinx/discussion/ forum]<br /> [https://github.com/cmusphinx github]<br /> |<br /> |Java (Sphinx4), C (pocketsphinx)<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |CMN, Mel-Spectrum subtraction<br /> |GMM, Streams<br /> |MLLR, MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch<br /> |{{yes|Yes}}<br /> |[http://www.keithv.com/software/sphinx/ AURORA4 (WSJ0)]<br /> |<br /> |-<br /> !HTK<br /> |1993-2009 (3.4.1) <br /> |{{yes|Yes}}<br /> |{{no|[http://htk.eng.cam.ac.uk/docs/license.shtml proprietary]}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://htk.eng.cam.ac.uk/download.shtml website]<br /> [http://htk.eng.cam.ac.uk/docs/docs.shtml book]<br /> [http://htk.eng.cam.ac.uk/mailing/subscribe_mail.shtml mail-list]<br /> |[http://htk.eng.cam.ac.uk/extensions/index.shtml official], [http://htk.eng.cam.ac.uk/develop/atk.shtml ATK], [https://github.com/ramon-astudillo/custom_fe FE uncertainty decoding]<br /> |C<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMN<br /> |GMM (Full Cov.), Tied-Mix, Streams<br /> |HLDA, MLLR (w/ reg. trees), CMLR (w/ adaptive training), MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch, MMI, MPE, MWE<br /> |{{yes|Yes}}<br /> |[http://catalog.elra.info/product_info.php?cPath=37_40&amp;products_id=693 AURORA2 (purch.)] [http://catalog.elra.info/index.php?cPath=37_40 AURORA3 (purch.)], [http://www.keithv.com/software/htk/ AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/projects/chime/PCC/data/pcchome.tar.gz CHIME-1], [ftp://ftp.dcs.shef.ac.uk/share/spandh/chime_challenge/grid/eval_tools_grid.tgz CHIME-2-I], [http://spandh.dcs.shef.ac.uk/chime_challenge/chime2_task2.html CHIME-2-II],[http://reverb2014.dereverberation.com/tools/REVERB_TOOLS_FOR_ASR_ver2.0.tgz REVERB]<br /> |ETSI-AFE-AURORA2 [http://aurora.hsnr.de/download/Aurora2_afe_v1_1.pdf paper] (see AURORA2 purch.)<br /> |-<br /> !Kaldi<br /> |2009-* (continous updates)<br /> |{{yes|Yes}}<br /> |{{Yes|Apache 2.0}}<br /> |{{some|Windows (not mantained as of 2014), Linux, OSX}}<br /> |[http://kaldi.sourceforge.net/about.html website]<br /> [http://homepages.inf.ed.ac.uk/aghoshal/pubs/asru11-kaldi.pdf paper]<br /> [http://sourceforge.net/p/kaldi/mailman/kaldi-users/ mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://kaldi.sourceforge.net/install.html SVN]<br /> |<br /> |C++<br /> |{{yes|BLAS, LAPACK, GPU (for DNNs)}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMVN<br /> |GMM (Full Cov.), SGMM, DNN<br /> |HLDA, STC, MLLT, MLLR, CMLLR (w/ reg. trees), Exponential transform<br /> |aligment, N-best, lattice rescoring (using OpenFST)<br /> |Baum-Welch, MMI (boosted), MC, feature-based<br /> |{{yes|Yes}}<br /> |[http://kaldi.sourceforge.net/data_prep.html AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/chime_challenge/WSJ0public/CHiME2012-WSJ0-Kaldi_0.03.tar.gz CHIME-2]<br /> |Weniger2014-REVERB [http://reverb2014.dereverberation.com/workshop/reverb2014-papers/1569884459.pdf Paper] [http://www.mmk.ei.tum.de/~wen/REVERB_2014/kaldi_baseline.tar.gz Code]<br /> |-<br /> !Spraak<br /> |2008-* (1.1.374)<br /> |{{yes|Yes}}<br /> |{{no|[http://www.spraak.org/obtaining-spraak/license proprietary2]}}<br /> |{{some|Windows ([http://www.spraak.org/obtaining-spraak/system-requirements limited]), Linux, OSX}}<br /> |[http://www.spraak.org/ website]<br /> [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/wambacq/interspeech08/is2008_spraak_v3.pdf paper]<br /> [http://www.spraak.org/mailing-lists mail-list]<br /> [http://www.spraak.org/mailing-lists mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://www.spraak.org/documentation/doxygen/doc/html/spr__svn.html SVN]<br /> |Missing Data Techniques (MDT)<br /> |C, Python<br /> |{{no|No}} <br /> |{{yes|Yes}}<br /> |Flexible preprocessing script language -- examples for MFCC, PLP <br /> |VTLN,CMN, [http://www.spraak.org/documentation/doxygen/doc/html/spr__tut__mida.html MIDA], [http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__intro.html MDT Techniques], Parametric HistEq [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/ICASSP2010/zhang.pdf&amp;auto&amp;xz:icassp10], Noise normalization [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |GMM (Tied-Mix), Exemplar based [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/dtw/paper_dtw.pdf&amp;auto&amp;kd:icassp11a], NN, CRF, ... (flexible using the preprocessing script) [http://lib.ugent.be/catalog/pug01:4382368]<br /> |CMLLR, eigenvoices, GMM-weight based (NMF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/SPEECOM2013/manuscript_zhang.pdf&amp;auto&amp;xzhang13] -- (all have Matlab dependencies); MAP<br /> |aligment, lattice rescoring, SCRF rescoring (using SCARF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/scarf/paper_scarf.pdf&amp;auto&amp;kd:icassp11b], phone lattice rescoring [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/duchato/eurospeech09/flavor.pdf&amp;auto&amp;jd:intersp09]<br /> |Viterbi<br /> |{{yes|Yes}<br /> |[http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__example.html AURORA4], [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |<br /> |}<br /> <br /> == [[Speaker identification and verification]] ==<br /> <br /> == [[Speech enhancement and separation]] ==<br /> <br /> == [[Other applications]] ==<br /> <br /> == Contribute software ==<br /> To contribute new software, please<br /> * [[Special:UserLogin|create an account]] and login<br /> * go to the wiki page above corresponding to your application; if it does not exist yet, you may create it<br /> * click on the &quot;Edit&quot; link at the top of the page and add a new section for your software (software is ordered by year of the latest version)<br /> * click on the &quot;Save page&quot; link at the bottom of the page to save your modifications<br /> <br /> Please make sure to provide the following information:<br /> * name of the software and year of the latest version<br /> * authors, institution, contact information<br /> * link to the software, ideally including a short demo, and to the external libraries needed<br /> * short description (functionalities, inputs and outputs, programming language, operating system, license, etc) and link to a paper/report describing the software, if any<br /> * whether running on well-known baselines (Aurora-2, Aurora-4, Switchboard, CHiME, etc) is included or requires wrapping by the user<br /> <br /> In order to save storage space, please do not upload the software on this wiki, but link it as much as possible from a public repository (e.g., bitbucket, github, sourceforge) or from a stable URL on the website of your institution. If this is not possible, please contact the [[Main_Page#Working group contacts|resources sharing working group]].</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=Software&diff=323 Software 2014-11-26T09:09:51Z <p>Ramon.astudillo: /* Automatic speech recognition */</p> <hr /> <div>This page provides software grouped by application. <br /> <br /> == [[Automatic speech recognition]] ==<br /> <br /> {| class=&quot;wikitable sortable&quot; style=&quot;font-size:72%; border:gray solid 1px; text-align:center; width:auto; table-layout:fixed;&quot;<br /> |-<br /> !style=&quot;width: 40px&quot; rowspan=&quot;2&quot; class=&quot;unsortable&quot;| ASR engines<br /> !colspan=&quot;6&quot; |General attributes<br /> !colspan=&quot;2&quot; |Programming<br /> !colspan=&quot;8&quot; |Implemented ASR techniques<br /> !colspan=&quot;2&quot; |Reproducible research<br /> |-<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | release / update<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | actively developed<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | licence<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | platforms<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | links <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | extensions<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | language<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | hardware optimization<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | VAD<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic features<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | feature normalization / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic models<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | model adaptation / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | decoding techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | training techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | online ASR<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | robust ASR training recipes <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | reproducible results<br /> |-<br /> !CMU Sphinx<br /> |1986-* (Sphinx 4.1.0, pocketsphinx 0.8) <br /> |{{yes|Yes}}<br /> |{{yes|[https://raw.githubusercontent.com/cmusphinx/sphinx4/master/license.terms BSD-like]}}<br /> |{{yes|Windows, Linux, OSX (Sphinx4) / Raspberry-pi, iPhone, Android (pocketsphinx)}}<br /> |[http://cmusphinx.sourceforge.net/ website] <br /> [http://www.cs.cmu.edu/~rsingh/homepage/papers/icassp03-sphinx4_2.pdf paper]<br /> [https://www.cs.cmu.edu/~awb/papers/ICASSP2006/0100185.pdf paper]<br /> [http://sourceforge.net/p/cmusphinx/mailman/ mail-list]<br /> [http://sourceforge.net/p/cmusphinx/discussion/ forum]<br /> [https://github.com/cmusphinx github]<br /> |<br /> |Java (Sphinx4), C (pocketsphinx)<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |CMN, Mel-Spectrum subtraction<br /> |GMM, Streams<br /> |MLLR, MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch<br /> |{{yes|Yes}}<br /> |[http://www.keithv.com/software/sphinx/ AURORA4 (WSJ0)]<br /> |<br /> |-<br /> !HTK<br /> |1993-2009 (3.4.1) <br /> |{{yes|Yes}}<br /> |{{no|[http://htk.eng.cam.ac.uk/docs/license.shtml proprietary]}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://htk.eng.cam.ac.uk/download.shtml website]<br /> [http://htk.eng.cam.ac.uk/docs/docs.shtml book]<br /> [http://htk.eng.cam.ac.uk/mailing/subscribe_mail.shtml mail-list]<br /> |[http://htk.eng.cam.ac.uk/extensions/index.shtml official], [http://htk.eng.cam.ac.uk/develop/atk.shtml ATK], [https://github.com/ramon-astudillo/custom_fe FE uncertainty decoding]<br /> |C<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMN<br /> |GMM (Full Cov.), Tied-Mix, Streams<br /> |HLDA, MLLR (w/ reg. trees), CMLR (w/ adaptive training), MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch, MMI, MPE, MWE<br /> |{{yes|Yes}}<br /> |[http://catalog.elra.info/product_info.php?cPath=37_40&amp;products_id=693 AURORA2 (purch.)] [http://catalog.elra.info/index.php?cPath=37_40 AURORA3 (purch.)], [http://www.keithv.com/software/htk/ AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/projects/chime/PCC/data/pcchome.tar.gz CHIME-1], [ftp://ftp.dcs.shef.ac.uk/share/spandh/chime_challenge/grid/eval_tools_grid.tgz CHIME-2-I], [http://spandh.dcs.shef.ac.uk/chime_challenge/chime2_task2.html CHIME-2-II],[http://reverb2014.dereverberation.com/tools/REVERB_TOOLS_FOR_ASR_ver2.0.tgz REVERB]<br /> |ETSI-AFE-AURORA2 [http://aurora.hsnr.de/download/Aurora2_afe_v1_1.pdf paper] (see AURORA2 purch.)<br /> |-<br /> !Kaldi<br /> |2009-* (continous updates)<br /> |{{yes|Yes}}<br /> |{{Yes|Apache 2.0}}<br /> |{{some|Windows (not mantained as of 2014), Linux, OSX}}<br /> |[http://kaldi.sourceforge.net/about.html website]<br /> [http://homepages.inf.ed.ac.uk/aghoshal/pubs/asru11-kaldi.pdf paper]<br /> [http://sourceforge.net/p/kaldi/mailman/kaldi-users/ mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://kaldi.sourceforge.net/install.html SVN]<br /> |<br /> |C++<br /> |{{yes|BLAS, LAPACK, GPU (for DNNs)}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMVN<br /> |GMM (Full Cov.), SGMM, DNN<br /> |HLDA, STC, MLLT, MLLR, CMLLR (w/ reg. trees), Exponential transform<br /> |aligment, N-best, lattice rescoring (using OpenFST)<br /> |Baum-Welch, MMI (boosted), MC, feature-based<br /> |{{yes|Yes}}<br /> |[http://kaldi.sourceforge.net/data_prep.html AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/chime_challenge/WSJ0public/CHiME2012-WSJ0-Kaldi_0.03.tar.gz CHIME-2]<br /> |Weniger2014-REVERB [http://reverb2014.dereverberation.com/workshop/reverb2014-papers/1569884459.pdf Paper] [http://www.mmk.ei.tum.de/~wen/REVERB_2014/kaldi_baseline.tar.gz Code]<br /> |-<br /> !Spraak<br /> |2008-* (1.1.374)<br /> |{{yes|Yes}}<br /> |{{no|[http://www.spraak.org/obtaining-spraak/license proprietary2]}}<br /> |{{some|Windows ([http://www.spraak.org/obtaining-spraak/system-requirements limited]), Linux, OSX}}<br /> |[http://www.spraak.org/ website]<br /> [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/wambacq/interspeech08/is2008_spraak_v3.pdf paper]<br /> [http://www.spraak.org/mailing-lists mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://www.spraak.org/documentation/doxygen/doc/html/spr__svn.html SVN]<br /> |Missing Data Techniques (MDT)<br /> |C, Python<br /> |{{no|No}} <br /> |{{yes|Yes}}<br /> |Flexible preprocessing script language -- examples for MFCC, PLP <br /> |VTLN,CMN, [http://www.spraak.org/documentation/doxygen/doc/html/spr__tut__mida.html MIDA], [http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__intro.html MDT Techniques], Parametric HistEq [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/ICASSP2010/zhang.pdf&amp;auto&amp;xz:icassp10], Noise normalization [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |GMM (Tied-Mix), Exemplar based [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/dtw/paper_dtw.pdf&amp;auto&amp;kd:icassp11a], NN, CRF, ... (flexible using the preprocessing script) [http://lib.ugent.be/catalog/pug01:4382368]<br /> |CMLLR, eigenvoices, GMM-weight based (NMF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/SPEECOM2013/manuscript_zhang.pdf&amp;auto&amp;xzhang13] -- (all have Matlab dependencies); MAP<br /> |aligment, lattice rescoring, SCRF rescoring (using SCARF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/scarf/paper_scarf.pdf&amp;auto&amp;kd:icassp11b], phone lattice rescoring [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/duchato/eurospeech09/flavor.pdf&amp;auto&amp;jd:intersp09]<br /> |Viterbi<br /> |{{yes|Yes}<br /> |[http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__example.html AURORA4], [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |<br /> |}<br /> <br /> == [[Speaker identification and verification]] ==<br /> <br /> == [[Speech enhancement and separation]] ==<br /> <br /> == [[Other applications]] ==<br /> <br /> == Contribute software ==<br /> To contribute new software, please<br /> * [[Special:UserLogin|create an account]] and login<br /> * go to the wiki page above corresponding to your application; if it does not exist yet, you may create it<br /> * click on the &quot;Edit&quot; link at the top of the page and add a new section for your software (software is ordered by year of the latest version)<br /> * click on the &quot;Save page&quot; link at the bottom of the page to save your modifications<br /> <br /> Please make sure to provide the following information:<br /> * name of the software and year of the latest version<br /> * authors, institution, contact information<br /> * link to the software, ideally including a short demo, and to the external libraries needed<br /> * short description (functionalities, inputs and outputs, programming language, operating system, license, etc) and link to a paper/report describing the software, if any<br /> * whether running on well-known baselines (Aurora-2, Aurora-4, Switchboard, CHiME, etc) is included or requires wrapping by the user<br /> <br /> In order to save storage space, please do not upload the software on this wiki, but link it as much as possible from a public repository (e.g., bitbucket, github, sourceforge) or from a stable URL on the website of your institution. If this is not possible, please contact the [[Main_Page#Working group contacts|resources sharing working group]].</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=Software&diff=322 Software 2014-11-26T09:09:02Z <p>Ramon.astudillo: /* Automatic speech recognition */</p> <hr /> <div>This page provides software grouped by application. <br /> <br /> == [[Automatic speech recognition]] ==<br /> <br /> {| class=&quot;wikitable sortable&quot; style=&quot;font-size:72%; border:gray solid 1px; text-align:center; width:auto; table-layout:fixed;&quot;<br /> |-<br /> !style=&quot;width: 40px&quot; rowspan=&quot;2&quot; class=&quot;unsortable&quot;| ASR engines<br /> !colspan=&quot;6&quot; |General attributes<br /> !colspan=&quot;2&quot; |Programming<br /> !colspan=&quot;8&quot; |Implemented ASR techniques<br /> !colspan=&quot;2&quot; |Reproducible research<br /> |-<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | release / update<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | actively developed<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | licence<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | platforms<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | links <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | extensions<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | language<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | hardware optimization<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | VAD<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic features<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | feature normalization / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic models<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | model adaptation / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | decoding techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | training techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | online ASR<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | robust ASR training recipes <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | reproducible results<br /> |-<br /> !CMU Sphinx<br /> |1986-* (Sphinx 4.1.0, pocketsphinx 0.8) <br /> |{{yes|Yes}}<br /> |{{yes|[https://raw.githubusercontent.com/cmusphinx/sphinx4/master/license.terms BSD-like]}}<br /> |{{yes|Windows, Linux, OSX (Sphinx4) / Raspberry-pi, iPhone, Android (pocketsphinx)}}<br /> |[http://cmusphinx.sourceforge.net/ website] <br /> [http://www.cs.cmu.edu/~rsingh/homepage/papers/icassp03-sphinx4_2.pdf paper]<br /> [https://www.cs.cmu.edu/~awb/papers/ICASSP2006/0100185.pdf paper]<br /> [http://sourceforge.net/p/cmusphinx/mailman/ mail-list]<br /> [http://sourceforge.net/p/cmusphinx/discussion/ forum]<br /> [https://github.com/cmusphinx github]<br /> |<br /> |Java (Sphinx4), C (pocketsphinx)<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |CMN, Mel-Spectrum subtraction<br /> |GMM, Streams<br /> |MLLR, MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch<br /> |{{yes|Yes}}<br /> |[http://www.keithv.com/software/sphinx/ AURORA4 (WSJ0)]<br /> |<br /> |-<br /> !HTK<br /> |1993-2009 (3.4.1) <br /> |{{yes|Yes}}<br /> |{{no|[http://htk.eng.cam.ac.uk/docs/license.shtml proprietary]}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://htk.eng.cam.ac.uk/download.shtml website]<br /> [http://htk.eng.cam.ac.uk/docs/docs.shtml book]<br /> [http://htk.eng.cam.ac.uk/mailing/subscribe_mail.shtml mail-list]<br /> |[http://htk.eng.cam.ac.uk/extensions/index.shtml official], [http://htk.eng.cam.ac.uk/develop/atk.shtml ATK], [https://github.com/ramon-astudillo/custom_fe FE uncertainty decoding]<br /> |C<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMN<br /> |GMM (Full Cov.), Tied-Mix, Streams<br /> |HLDA, MLLR (w/ reg. trees), CMLR (w/ adaptive training), MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch, MMI, MPE, MWE<br /> |{{yes|Yes}}<br /> |[http://catalog.elra.info/product_info.php?cPath=37_40&amp;products_id=693 AURORA2 (purch.)] [http://catalog.elra.info/index.php?cPath=37_40 AURORA3 (purch.)], [http://www.keithv.com/software/htk/ AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/projects/chime/PCC/data/pcchome.tar.gz CHIME-1], [ftp://ftp.dcs.shef.ac.uk/share/spandh/chime_challenge/grid/eval_tools_grid.tgz CHIME-2-I], [http://spandh.dcs.shef.ac.uk/chime_challenge/chime2_task2.html CHIME-2-II],[http://reverb2014.dereverberation.com/tools/REVERB_TOOLS_FOR_ASR_ver2.0.tgz REVERB]<br /> |ETSI-AFE-AURORA2 [http://aurora.hsnr.de/download/Aurora2_afe_v1_1.pdf paper] (see AURORA2 purch.)<br /> |-<br /> !Kaldi<br /> |2009-* (continous updates)<br /> |{{yes|Yes}}<br /> |{{Yes|Apache 2.0}}<br /> |{{some|Windows (not mantained as of 2014), Linux, OSX}}<br /> |[http://kaldi.sourceforge.net/about.html website]<br /> [http://homepages.inf.ed.ac.uk/aghoshal/pubs/asru11-kaldi.pdf paper]<br /> [http://sourceforge.net/p/kaldi/mailman/kaldi-users/ mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://kaldi.sourceforge.net/install.html SVN]<br /> |<br /> |C++<br /> |{{yes|BLAS, LAPACK, GPU (for DNNs)}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMVN<br /> |GMM (Full Cov.), SGMM, DNN<br /> |HLDA, STC, MLLT, MLLR, CMLLR (w/ reg. trees), Exponential transform<br /> |aligment, N-best, lattice rescoring (using OpenFST)<br /> |Baum-Welch, MMI (boosted), MC, feature-based<br /> |{{yes|Yes}}<br /> |[http://kaldi.sourceforge.net/data_prep.html AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/chime_challenge/WSJ0public/CHiME2012-WSJ0-Kaldi_0.03.tar.gz CHIME-2]<br /> |Weniger2014-REVERB [http://reverb2014.dereverberation.com/workshop/reverb2014-papers/1569884459.pdf Paper] [http://www.mmk.ei.tum.de/~wen/REVERB_2014/kaldi_baseline.tar.gz Code]<br /> [http://sourceforge.net/p/kaldi/discussion/ Kaldi]<br /> |-<br /> !Spraak<br /> |2008-* (1.1.374)<br /> |{{yes|Yes}}<br /> |{{no|[http://www.spraak.org/obtaining-spraak/license proprietary2]}}<br /> |{{some|Windows ([http://www.spraak.org/obtaining-spraak/system-requirements limited]), Linux, OSX}}<br /> |[http://www.spraak.org/ website]<br /> [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/wambacq/interspeech08/is2008_spraak_v3.pdf paper]<br /> [http://www.spraak.org/mailing-lists mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://www.spraak.org/documentation/doxygen/doc/html/spr__svn.html SVN]<br /> |Missing Data Techniques (MDT)<br /> |C, Python<br /> |{{no|No}} <br /> |{{yes|Yes}}<br /> |Flexible preprocessing script language -- examples for MFCC, PLP <br /> |VTLN,CMN, [http://www.spraak.org/documentation/doxygen/doc/html/spr__tut__mida.html MIDA], [http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__intro.html MDT Techniques], Parametric HistEq [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/ICASSP2010/zhang.pdf&amp;auto&amp;xz:icassp10], Noise normalization [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |GMM (Tied-Mix), Exemplar based [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/dtw/paper_dtw.pdf&amp;auto&amp;kd:icassp11a], NN, CRF, ... (flexible using the preprocessing script) [http://lib.ugent.be/catalog/pug01:4382368]<br /> |CMLLR, eigenvoices, GMM-weight based (NMF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/SPEECOM2013/manuscript_zhang.pdf&amp;auto&amp;xzhang13] -- (all have Matlab dependencies); MAP<br /> |aligment, lattice rescoring, SCRF rescoring (using SCARF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/scarf/paper_scarf.pdf&amp;auto&amp;kd:icassp11b], phone lattice rescoring [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/duchato/eurospeech09/flavor.pdf&amp;auto&amp;jd:intersp09]<br /> |Viterbi<br /> |{{yes|Yes}<br /> |[http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__example.html AURORA4], [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |<br /> |}<br /> <br /> == [[Speaker identification and verification]] ==<br /> <br /> == [[Speech enhancement and separation]] ==<br /> <br /> == [[Other applications]] ==<br /> <br /> == Contribute software ==<br /> To contribute new software, please<br /> * [[Special:UserLogin|create an account]] and login<br /> * go to the wiki page above corresponding to your application; if it does not exist yet, you may create it<br /> * click on the &quot;Edit&quot; link at the top of the page and add a new section for your software (software is ordered by year of the latest version)<br /> * click on the &quot;Save page&quot; link at the bottom of the page to save your modifications<br /> <br /> Please make sure to provide the following information:<br /> * name of the software and year of the latest version<br /> * authors, institution, contact information<br /> * link to the software, ideally including a short demo, and to the external libraries needed<br /> * short description (functionalities, inputs and outputs, programming language, operating system, license, etc) and link to a paper/report describing the software, if any<br /> * whether running on well-known baselines (Aurora-2, Aurora-4, Switchboard, CHiME, etc) is included or requires wrapping by the user<br /> <br /> In order to save storage space, please do not upload the software on this wiki, but link it as much as possible from a public repository (e.g., bitbucket, github, sourceforge) or from a stable URL on the website of your institution. If this is not possible, please contact the [[Main_Page#Working group contacts|resources sharing working group]].</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=Software&diff=321 Software 2014-11-26T09:08:43Z <p>Ramon.astudillo: /* Other applications */</p> <hr /> <div>This page provides software grouped by application. <br /> <br /> == [[Automatic speech recognition]] ==<br /> <br /> {| class=&quot;wikitable sortable&quot; style=&quot;font-size:72%; border:gray solid 1px; text-align:center; width:auto; table-layout:fixed;&quot;<br /> |-<br /> !style=&quot;width: 40px&quot; rowspan=&quot;2&quot; class=&quot;unsortable&quot;| ASR engines<br /> !colspan=&quot;6&quot; |General attributes<br /> !colspan=&quot;2&quot; |Programming<br /> !colspan=&quot;8&quot; |Implemented ASR techniques<br /> !colspan=&quot;2&quot; |Reproducible research<br /> |-<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | release / update<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | actively developed<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | licence<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | platforms<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | links <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | extensions<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | language<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | hardware optimization<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | VAD<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic features<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | feature normalization / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic models<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | model adaptation / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | decoding techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | training techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | online ASR<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | robust ASR training recipes <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | reproducible results<br /> |-<br /> !CMU Sphinx<br /> |1986-* (Sphinx 4.1.0, pocketsphinx 0.8) <br /> |{{yes|Yes}}<br /> |{{yes|[https://raw.githubusercontent.com/cmusphinx/sphinx4/master/license.terms BSD-like]}}<br /> |{{yes|Windows, Linux, OSX (Sphinx4) / Raspberry-pi, iPhone, Android (pocketsphinx)}}<br /> |[http://cmusphinx.sourceforge.net/ website] <br /> [http://www.cs.cmu.edu/~rsingh/homepage/papers/icassp03-sphinx4_2.pdf paper]<br /> [https://www.cs.cmu.edu/~awb/papers/ICASSP2006/0100185.pdf paper]<br /> [http://sourceforge.net/p/cmusphinx/mailman/ mail-list]<br /> [http://sourceforge.net/p/cmusphinx/discussion/ forum]<br /> [https://github.com/cmusphinx github]<br /> |<br /> |Java (Sphinx4), C (pocketsphinx)<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |CMN, Mel-Spectrum subtraction<br /> |GMM, Streams<br /> |MLLR, MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch<br /> |{{yes|Yes}}<br /> |[http://www.keithv.com/software/sphinx/ AURORA4 (WSJ0)]<br /> |<br /> |-<br /> !HTK<br /> |1993-2009 (3.4.1) <br /> |{{yes|Yes}}<br /> |{{no|[http://htk.eng.cam.ac.uk/docs/license.shtml proprietary]}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://htk.eng.cam.ac.uk/download.shtml website]<br /> [http://htk.eng.cam.ac.uk/docs/docs.shtml book]<br /> [http://htk.eng.cam.ac.uk/mailing/subscribe_mail.shtml mail-list]<br /> |[http://htk.eng.cam.ac.uk/extensions/index.shtml official], [http://htk.eng.cam.ac.uk/develop/atk.shtml ATK], [https://github.com/ramon-astudillo/custom_fe FE uncertainty decoding]<br /> |C<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMN<br /> |GMM (Full Cov.), Tied-Mix, Streams<br /> |HLDA, MLLR (w/ reg. trees), CMLR (w/ adaptive training), MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch, MMI, MPE, MWE<br /> |{{yes|Yes}}<br /> |[http://catalog.elra.info/product_info.php?cPath=37_40&amp;products_id=693 AURORA2 (purch.)] [http://catalog.elra.info/index.php?cPath=37_40 AURORA3 (purch.)], [http://www.keithv.com/software/htk/ AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/projects/chime/PCC/data/pcchome.tar.gz CHIME-1], [ftp://ftp.dcs.shef.ac.uk/share/spandh/chime_challenge/grid/eval_tools_grid.tgz CHIME-2-I], [http://spandh.dcs.shef.ac.uk/chime_challenge/chime2_task2.html CHIME-2-II],[http://reverb2014.dereverberation.com/tools/REVERB_TOOLS_FOR_ASR_ver2.0.tgz REVERB]<br /> |ETSI-AFE-AURORA2 [http://aurora.hsnr.de/download/Aurora2_afe_v1_1.pdf paper] (see AURORA2 purch.)<br /> |-<br /> !Kaldi<br /> |2009-* (continous updates)<br /> |{{yes|Yes}}<br /> |{{Yes|Apache 2.0}}<br /> |{{some|Windows (not mantained as of 2014), Linux, OSX}}<br /> |[http://kaldi.sourceforge.net/about.html website]<br /> [http://homepages.inf.ed.ac.uk/aghoshal/pubs/asru11-kaldi.pdf paper]<br /> [http://sourceforge.net/p/kaldi/mailman/kaldi-users/ mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://kaldi.sourceforge.net/install.html SVN]<br /> |<br /> |C++<br /> |{{yes|BLAS, LAPACK, GPU (for DNNs)}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMVN<br /> |GMM (Full Cov.), SGMM, DNN<br /> |HLDA, STC, MLLT, MLLR, CMLLR (w/ reg. trees), Exponential transform<br /> |aligment, N-best, lattice rescoring (using OpenFST)<br /> |Baum-Welch, MMI (boosted), MC, feature-based<br /> |{{yes|Yes}}<br /> |[http://kaldi.sourceforge.net/data_prep.html AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/chime_challenge/WSJ0public/CHiME2012-WSJ0-Kaldi_0.03.tar.gz CHIME-2]<br /> |Weniger2014-REVERB [http://reverb2014.dereverberation.com/workshop/reverb2014-papers/1569884459.pdf Paper] [http://www.mmk.ei.tum.de/~wen/REVERB_2014/kaldi_baseline.tar.gz Code]<br /> |-<br /> !Spraak<br /> |2008-* (1.1.374)<br /> |{{yes|Yes}}<br /> |{{no|[http://www.spraak.org/obtaining-spraak/license proprietary2]}}<br /> |{{some|Windows ([http://www.spraak.org/obtaining-spraak/system-requirements limited]), Linux, OSX}}<br /> |[http://www.spraak.org/ website]<br /> [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/wambacq/interspeech08/is2008_spraak_v3.pdf paper]<br /> [http://www.spraak.org/mailing-lists mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://www.spraak.org/documentation/doxygen/doc/html/spr__svn.html SVN]<br /> |Missing Data Techniques (MDT)<br /> |C, Python<br /> |{{no|No}} <br /> |{{yes|Yes}}<br /> |Flexible preprocessing script language -- examples for MFCC, PLP <br /> |VTLN,CMN, [http://www.spraak.org/documentation/doxygen/doc/html/spr__tut__mida.html MIDA], [http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__intro.html MDT Techniques], Parametric HistEq [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/ICASSP2010/zhang.pdf&amp;auto&amp;xz:icassp10], Noise normalization [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |GMM (Tied-Mix), Exemplar based [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/dtw/paper_dtw.pdf&amp;auto&amp;kd:icassp11a], NN, CRF, ... (flexible using the preprocessing script) [http://lib.ugent.be/catalog/pug01:4382368]<br /> |CMLLR, eigenvoices, GMM-weight based (NMF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/SPEECOM2013/manuscript_zhang.pdf&amp;auto&amp;xzhang13] -- (all have Matlab dependencies); MAP<br /> |aligment, lattice rescoring, SCRF rescoring (using SCARF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/scarf/paper_scarf.pdf&amp;auto&amp;kd:icassp11b], phone lattice rescoring [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/duchato/eurospeech09/flavor.pdf&amp;auto&amp;jd:intersp09]<br /> |Viterbi<br /> |{{yes|Yes}<br /> |[http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__example.html AURORA4], [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |<br /> |}<br /> <br /> == [[Speaker identification and verification]] ==<br /> <br /> == [[Speech enhancement and separation]] ==<br /> <br /> == [[Other applications]] ==<br /> <br /> == Contribute software ==<br /> To contribute new software, please<br /> * [[Special:UserLogin|create an account]] and login<br /> * go to the wiki page above corresponding to your application; if it does not exist yet, you may create it<br /> * click on the &quot;Edit&quot; link at the top of the page and add a new section for your software (software is ordered by year of the latest version)<br /> * click on the &quot;Save page&quot; link at the bottom of the page to save your modifications<br /> <br /> Please make sure to provide the following information:<br /> * name of the software and year of the latest version<br /> * authors, institution, contact information<br /> * link to the software, ideally including a short demo, and to the external libraries needed<br /> * short description (functionalities, inputs and outputs, programming language, operating system, license, etc) and link to a paper/report describing the software, if any<br /> * whether running on well-known baselines (Aurora-2, Aurora-4, Switchboard, CHiME, etc) is included or requires wrapping by the user<br /> <br /> In order to save storage space, please do not upload the software on this wiki, but link it as much as possible from a public repository (e.g., bitbucket, github, sourceforge) or from a stable URL on the website of your institution. If this is not possible, please contact the [[Main_Page#Working group contacts|resources sharing working group]].</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=Software&diff=320 Software 2014-11-26T09:08:17Z <p>Ramon.astudillo: /* Other applications */</p> <hr /> <div>This page provides software grouped by application. <br /> <br /> == [[Automatic speech recognition]] ==<br /> <br /> {| class=&quot;wikitable sortable&quot; style=&quot;font-size:72%; border:gray solid 1px; text-align:center; width:auto; table-layout:fixed;&quot;<br /> |-<br /> !style=&quot;width: 40px&quot; rowspan=&quot;2&quot; class=&quot;unsortable&quot;| ASR engines<br /> !colspan=&quot;6&quot; |General attributes<br /> !colspan=&quot;2&quot; |Programming<br /> !colspan=&quot;8&quot; |Implemented ASR techniques<br /> !colspan=&quot;2&quot; |Reproducible research<br /> |-<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | release / update<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | actively developed<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | licence<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | platforms<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | links <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | extensions<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | language<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | hardware optimization<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | VAD<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic features<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | feature normalization / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic models<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | model adaptation / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | decoding techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | training techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | online ASR<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | robust ASR training recipes <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | reproducible results<br /> |-<br /> !CMU Sphinx<br /> |1986-* (Sphinx 4.1.0, pocketsphinx 0.8) <br /> |{{yes|Yes}}<br /> |{{yes|[https://raw.githubusercontent.com/cmusphinx/sphinx4/master/license.terms BSD-like]}}<br /> |{{yes|Windows, Linux, OSX (Sphinx4) / Raspberry-pi, iPhone, Android (pocketsphinx)}}<br /> |[http://cmusphinx.sourceforge.net/ website] <br /> [http://www.cs.cmu.edu/~rsingh/homepage/papers/icassp03-sphinx4_2.pdf paper]<br /> [https://www.cs.cmu.edu/~awb/papers/ICASSP2006/0100185.pdf paper]<br /> [http://sourceforge.net/p/cmusphinx/mailman/ mail-list]<br /> [http://sourceforge.net/p/cmusphinx/discussion/ forum]<br /> [https://github.com/cmusphinx github]<br /> |<br /> |Java (Sphinx4), C (pocketsphinx)<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |CMN, Mel-Spectrum subtraction<br /> |GMM, Streams<br /> |MLLR, MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch<br /> |{{yes|Yes}}<br /> |[http://www.keithv.com/software/sphinx/ AURORA4 (WSJ0)]<br /> |<br /> |-<br /> !HTK<br /> |1993-2009 (3.4.1) <br /> |{{yes|Yes}}<br /> |{{no|[http://htk.eng.cam.ac.uk/docs/license.shtml proprietary]}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://htk.eng.cam.ac.uk/download.shtml website]<br /> [http://htk.eng.cam.ac.uk/docs/docs.shtml book]<br /> [http://htk.eng.cam.ac.uk/mailing/subscribe_mail.shtml mail-list]<br /> |[http://htk.eng.cam.ac.uk/extensions/index.shtml official], [http://htk.eng.cam.ac.uk/develop/atk.shtml ATK], [https://github.com/ramon-astudillo/custom_fe FE uncertainty decoding]<br /> |C<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMN<br /> |GMM (Full Cov.), Tied-Mix, Streams<br /> |HLDA, MLLR (w/ reg. trees), CMLR (w/ adaptive training), MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch, MMI, MPE, MWE<br /> |{{yes|Yes}}<br /> |[http://catalog.elra.info/product_info.php?cPath=37_40&amp;products_id=693 AURORA2 (purch.)] [http://catalog.elra.info/index.php?cPath=37_40 AURORA3 (purch.)], [http://www.keithv.com/software/htk/ AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/projects/chime/PCC/data/pcchome.tar.gz CHIME-1], [ftp://ftp.dcs.shef.ac.uk/share/spandh/chime_challenge/grid/eval_tools_grid.tgz CHIME-2-I], [http://spandh.dcs.shef.ac.uk/chime_challenge/chime2_task2.html CHIME-2-II],[http://reverb2014.dereverberation.com/tools/REVERB_TOOLS_FOR_ASR_ver2.0.tgz REVERB]<br /> |ETSI-AFE-AURORA2 [http://aurora.hsnr.de/download/Aurora2_afe_v1_1.pdf paper] (see AURORA2 purch.)<br /> |-<br /> !Kaldi<br /> |2009-* (continous updates)<br /> |{{yes|Yes}}<br /> |{{Yes|Apache 2.0}}<br /> |{{some|Windows (not mantained as of 2014), Linux, OSX}}<br /> |[http://kaldi.sourceforge.net/about.html website]<br /> [http://homepages.inf.ed.ac.uk/aghoshal/pubs/asru11-kaldi.pdf paper]<br /> [http://sourceforge.net/p/kaldi/mailman/kaldi-users/ mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://kaldi.sourceforge.net/install.html SVN]<br /> |<br /> |C++<br /> |{{yes|BLAS, LAPACK, GPU (for DNNs)}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMVN<br /> |GMM (Full Cov.), SGMM, DNN<br /> |HLDA, STC, MLLT, MLLR, CMLLR (w/ reg. trees), Exponential transform<br /> |aligment, N-best, lattice rescoring (using OpenFST)<br /> |Baum-Welch, MMI (boosted), MC, feature-based<br /> |{{yes|Yes}}<br /> |[http://kaldi.sourceforge.net/data_prep.html AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/chime_challenge/WSJ0public/CHiME2012-WSJ0-Kaldi_0.03.tar.gz CHIME-2]<br /> |Weniger2014-REVERB [http://reverb2014.dereverberation.com/workshop/reverb2014-papers/1569884459.pdf Paper] [http://www.mmk.ei.tum.de/~wen/REVERB_2014/kaldi_baseline.tar.gz Code]<br /> |-<br /> !Spraak<br /> |2008-* (1.1.374)<br /> |{{yes|Yes}}<br /> |{{no|[http://www.spraak.org/obtaining-spraak/license proprietary2]}}<br /> |{{some|Windows ([http://www.spraak.org/obtaining-spraak/system-requirements limited]), Linux, OSX}}<br /> |[http://www.spraak.org/ website]<br /> [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/wambacq/interspeech08/is2008_spraak_v3.pdf paper]<br /> [http://www.spraak.org/mailing-lists mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://www.spraak.org/documentation/doxygen/doc/html/spr__svn.html SVN]<br /> |Missing Data Techniques (MDT)<br /> |C, Python<br /> |{{no|No}} <br /> |{{yes|Yes}}<br /> |Flexible preprocessing script language -- examples for MFCC, PLP <br /> |VTLN,CMN, [http://www.spraak.org/documentation/doxygen/doc/html/spr__tut__mida.html MIDA], [http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__intro.html MDT Techniques], Parametric HistEq [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/ICASSP2010/zhang.pdf&amp;auto&amp;xz:icassp10], Noise normalization [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |GMM (Tied-Mix), Exemplar based [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/dtw/paper_dtw.pdf&amp;auto&amp;kd:icassp11a], NN, CRF, ... (flexible using the preprocessing script) [http://lib.ugent.be/catalog/pug01:4382368]<br /> |CMLLR, eigenvoices, GMM-weight based (NMF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/SPEECOM2013/manuscript_zhang.pdf&amp;auto&amp;xzhang13] -- (all have Matlab dependencies); MAP<br /> |aligment, lattice rescoring, SCRF rescoring (using SCARF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/scarf/paper_scarf.pdf&amp;auto&amp;kd:icassp11b], phone lattice rescoring [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/duchato/eurospeech09/flavor.pdf&amp;auto&amp;jd:intersp09]<br /> |Viterbi<br /> |{{yes|Yes}<br /> |[http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__example.html AURORA4], [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |<br /> |}<br /> <br /> == [[Speaker identification and verification]] ==<br /> <br /> == [[Speech enhancement and separation]] ==<br /> <br /> == [[Other applications]] ==<br /> http://sourceforge.net/p/kaldi/discussion/<br /> <br /> == Contribute software ==<br /> To contribute new software, please<br /> * [[Special:UserLogin|create an account]] and login<br /> * go to the wiki page above corresponding to your application; if it does not exist yet, you may create it<br /> * click on the &quot;Edit&quot; link at the top of the page and add a new section for your software (software is ordered by year of the latest version)<br /> * click on the &quot;Save page&quot; link at the bottom of the page to save your modifications<br /> <br /> Please make sure to provide the following information:<br /> * name of the software and year of the latest version<br /> * authors, institution, contact information<br /> * link to the software, ideally including a short demo, and to the external libraries needed<br /> * short description (functionalities, inputs and outputs, programming language, operating system, license, etc) and link to a paper/report describing the software, if any<br /> * whether running on well-known baselines (Aurora-2, Aurora-4, Switchboard, CHiME, etc) is included or requires wrapping by the user<br /> <br /> In order to save storage space, please do not upload the software on this wiki, but link it as much as possible from a public repository (e.g., bitbucket, github, sourceforge) or from a stable URL on the website of your institution. If this is not possible, please contact the [[Main_Page#Working group contacts|resources sharing working group]].</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=Software&diff=314 Software 2014-11-17T19:04:24Z <p>Ramon.astudillo: /* Automatic speech recognition */</p> <hr /> <div>This page provides software grouped by application. <br /> <br /> == [[Automatic speech recognition]] ==<br /> <br /> {| class=&quot;wikitable sortable&quot; style=&quot;font-size:72%; border:gray solid 1px; text-align:center; width:auto; table-layout:fixed;&quot;<br /> |-<br /> !style=&quot;width: 40px&quot; rowspan=&quot;2&quot; class=&quot;unsortable&quot;| ASR engines<br /> !colspan=&quot;6&quot; |General attributes<br /> !colspan=&quot;2&quot; |Programming<br /> !colspan=&quot;8&quot; |Implemented ASR techniques<br /> !colspan=&quot;2&quot; |Reproducible research<br /> |-<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | release / update<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | actively developed<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | licence<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | platforms<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | links <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | extensions<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | language<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | hardware optimization<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | VAD<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic features<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | feature normalization / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic models<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | model adaptation / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | decoding techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | training techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | online ASR<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | robust ASR training recipes <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | reproducible results<br /> |-<br /> !CMU Sphinx<br /> |1986-* (Sphinx 4.1.0, pocketsphinx 0.8) <br /> |{{yes|Yes}}<br /> |{{yes|[https://raw.githubusercontent.com/cmusphinx/sphinx4/master/license.terms BSD-like]}}<br /> |{{yes|Windows, Linux, OSX (Sphinx4) / Raspberry-pi, iPhone, Android (pocketsphinx)}}<br /> |[http://cmusphinx.sourceforge.net/ website] <br /> [http://www.cs.cmu.edu/~rsingh/homepage/papers/icassp03-sphinx4_2.pdf paper]<br /> [https://www.cs.cmu.edu/~awb/papers/ICASSP2006/0100185.pdf paper]<br /> [http://sourceforge.net/p/cmusphinx/mailman/ mail-list]<br /> [http://sourceforge.net/p/cmusphinx/discussion/ forum]<br /> [https://github.com/cmusphinx github]<br /> |<br /> |Java (Sphinx4), C (pocketsphinx)<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |CMN, Mel-Spectrum subtraction<br /> |GMM, Streams<br /> |MLLR, MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch<br /> |{{yes|Yes}}<br /> |[http://www.keithv.com/software/sphinx/ AURORA4 (WSJ0)]<br /> |<br /> |-<br /> !HTK<br /> |1993-2009 (3.4.1) <br /> |{{yes|Yes}}<br /> |{{no|[http://htk.eng.cam.ac.uk/docs/license.shtml proprietary]}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://htk.eng.cam.ac.uk/download.shtml website]<br /> [http://htk.eng.cam.ac.uk/docs/docs.shtml book]<br /> [http://htk.eng.cam.ac.uk/mailing/subscribe_mail.shtml mail-list]<br /> |[http://htk.eng.cam.ac.uk/extensions/index.shtml official], [http://htk.eng.cam.ac.uk/develop/atk.shtml ATK], [https://github.com/ramon-astudillo/custom_fe FE uncertainty decoding]<br /> |C<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMN<br /> |GMM (Full Cov.), Tied-Mix, Streams<br /> |HLDA, MLLR (w/ reg. trees), CMLR (w/ adaptive training), MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch, MMI, MPE, MWE<br /> |{{yes|Yes}}<br /> |[http://catalog.elra.info/product_info.php?cPath=37_40&amp;products_id=693 AURORA2 (purch.)] [http://catalog.elra.info/index.php?cPath=37_40 AURORA3 (purch.)], [http://www.keithv.com/software/htk/ AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/projects/chime/PCC/data/pcchome.tar.gz CHIME-1], [ftp://ftp.dcs.shef.ac.uk/share/spandh/chime_challenge/grid/eval_tools_grid.tgz CHIME-2-I], [http://spandh.dcs.shef.ac.uk/chime_challenge/chime2_task2.html CHIME-2-II],[http://reverb2014.dereverberation.com/tools/REVERB_TOOLS_FOR_ASR_ver2.0.tgz REVERB]<br /> |ETSI-AFE-AURORA2 [http://aurora.hsnr.de/download/Aurora2_afe_v1_1.pdf paper] (see AURORA2 purch.)<br /> |-<br /> !Kaldi<br /> |2009-* (continous updates)<br /> |{{yes|Yes}}<br /> |{{Yes|Apache 2.0}}<br /> |{{some|Windows (not mantained as of 2014), Linux, OSX}}<br /> |[http://kaldi.sourceforge.net/about.html website]<br /> [http://homepages.inf.ed.ac.uk/aghoshal/pubs/asru11-kaldi.pdf paper]<br /> [http://sourceforge.net/p/kaldi/mailman/kaldi-users/ mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://kaldi.sourceforge.net/install.html SVN]<br /> |<br /> |C++<br /> |{{yes|BLAS, LAPACK, GPU (for DNNs)}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMVN<br /> |GMM (Full Cov.), SGMM, DNN<br /> |HLDA, STC, MLLT, MLLR, CMLLR (w/ reg. trees), Exponential transform<br /> |aligment, N-best, lattice rescoring (using OpenFST)<br /> |Baum-Welch, MMI (boosted), MC, feature-based<br /> |{{yes|Yes}}<br /> |[http://kaldi.sourceforge.net/data_prep.html AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/chime_challenge/WSJ0public/CHiME2012-WSJ0-Kaldi_0.03.tar.gz CHIME-2]<br /> |Weniger2014-REVERB [http://reverb2014.dereverberation.com/workshop/reverb2014-papers/1569884459.pdf Paper] [http://www.mmk.ei.tum.de/~wen/REVERB_2014/kaldi_baseline.tar.gz Code]<br /> |-<br /> !Spraak<br /> |2008-* (1.1.374)<br /> |{{yes|Yes}}<br /> |{{no|[http://www.spraak.org/obtaining-spraak/license proprietary2]}}<br /> |{{some|Windows ([http://www.spraak.org/obtaining-spraak/system-requirements limited]), Linux, OSX}}<br /> |[http://www.spraak.org/ website]<br /> [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/wambacq/interspeech08/is2008_spraak_v3.pdf paper]<br /> [http://www.spraak.org/mailing-lists mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://www.spraak.org/documentation/doxygen/doc/html/spr__svn.html SVN]<br /> |Missing Data Techniques (MDT)<br /> |C, Python<br /> |{{no|No}} <br /> |{{yes|Yes}}<br /> |Flexible preprocessing script language -- examples for MFCC, PLP <br /> |VTLN,CMN, [http://www.spraak.org/documentation/doxygen/doc/html/spr__tut__mida.html MIDA], [http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__intro.html MDT Techniques], Parametric HistEq [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/ICASSP2010/zhang.pdf&amp;auto&amp;xz:icassp10], Noise normalization [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |GMM (Tied-Mix), Exemplar based [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/dtw/paper_dtw.pdf&amp;auto&amp;kd:icassp11a], NN, CRF, ... (flexible using the preprocessing script) [http://lib.ugent.be/catalog/pug01:4382368]<br /> |CMLLR, eigenvoices, GMM-weight based (NMF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/SPEECOM2013/manuscript_zhang.pdf&amp;auto&amp;xzhang13] -- (all have Matlab dependencies); MAP<br /> |aligment, lattice rescoring, SCRF rescoring (using SCARF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/scarf/paper_scarf.pdf&amp;auto&amp;kd:icassp11b], phone lattice rescoring [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/duchato/eurospeech09/flavor.pdf&amp;auto&amp;jd:intersp09]<br /> |Viterbi<br /> |{{yes|Yes}<br /> |[http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__example.html AURORA4], [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |<br /> |}<br /> <br /> == [[Speaker identification and verification]] ==<br /> <br /> == [[Speech enhancement and separation]] ==<br /> <br /> == [[Other applications]] ==<br /> <br /> == Contribute software ==<br /> To contribute new software, please<br /> * [[Special:UserLogin|create an account]] and login<br /> * go to the wiki page above corresponding to your application; if it does not exist yet, you may create it<br /> * click on the &quot;Edit&quot; link at the top of the page and add a new section for your software (software is ordered by year of the latest version)<br /> * click on the &quot;Save page&quot; link at the bottom of the page to save your modifications<br /> <br /> Please make sure to provide the following information:<br /> * name of the software and year of the latest version<br /> * authors, institution, contact information<br /> * link to the software, ideally including a short demo, and to the external libraries needed<br /> * short description (functionalities, inputs and outputs, programming language, operating system, license, etc) and link to a paper/report describing the software, if any<br /> * whether running on well-known baselines (Aurora-2, Aurora-4, Switchboard, CHiME, etc) is included or requires wrapping by the user<br /> <br /> In order to save storage space, please do not upload the software on this wiki, but link it as much as possible from a public repository (e.g., bitbucket, github, sourceforge) or from a stable URL on the website of your institution. If this is not possible, please contact the [[Main_Page#Working group contacts|resources sharing working group]].</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=Software&diff=313 Software 2014-11-17T19:00:58Z <p>Ramon.astudillo: /* Automatic speech recognition */</p> <hr /> <div>This page provides software grouped by application. <br /> <br /> == [[Automatic speech recognition]] ==<br /> <br /> {| class=&quot;wikitable sortable&quot; style=&quot;font-size:72%; border:gray solid 1px; text-align:center; width:auto; table-layout:fixed;&quot;<br /> |-<br /> !style=&quot;width: 40px&quot; rowspan=&quot;2&quot; class=&quot;unsortable&quot;| ASR engines<br /> !colspan=&quot;6&quot; |General attributes<br /> !colspan=&quot;2&quot; |Programming<br /> !colspan=&quot;8&quot; |Implemented ASR techniques<br /> !colspan=&quot;2&quot; |Reproducible research<br /> |-<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | release / update<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | actively developed<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | licence<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | platforms<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | links <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | extensions<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | language<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | hardware optimization<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | VAD<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic features<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | feature normalization / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic models<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | model adaptation / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | decoding techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | training techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | online ASR<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | robust ASR training recipes <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | reproducible results<br /> |-<br /> !CMU Sphinx<br /> |1986-* (Sphinx 4.1.0, pocketsphinx 0.8) <br /> |{{yes|Yes}}<br /> |{{yes|[https://raw.githubusercontent.com/cmusphinx/sphinx4/master/license.terms BSD-like]}}<br /> |{{yes|Windows, Linux, OSX (Sphinx4) / Raspberry-pi, iPhone, Android (pocketsphinx)}}<br /> |[http://cmusphinx.sourceforge.net/ website] <br /> [http://www.cs.cmu.edu/~rsingh/homepage/papers/icassp03-sphinx4_2.pdf paper]<br /> [https://www.cs.cmu.edu/~awb/papers/ICASSP2006/0100185.pdf paper]<br /> [http://sourceforge.net/p/cmusphinx/mailman/ mail-list]<br /> [http://sourceforge.net/p/cmusphinx/discussion/ forum]<br /> [https://github.com/cmusphinx github]<br /> |<br /> |Java (Sphinx4), C (pocketsphinx)<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |CMN, Mel-Spectrum subtraction<br /> |GMM, Streams<br /> |MLLR, MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch<br /> |{{yes|Yes}}<br /> |[http://www.keithv.com/software/sphinx/ AURORA4 (WSJ0)]<br /> |<br /> |-<br /> !HTK<br /> |1993-2009 (3.4.1) <br /> |{{yes|Yes}}<br /> |{{no|[http://htk.eng.cam.ac.uk/docs/license.shtml proprietary]}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://htk.eng.cam.ac.uk/download.shtml website]<br /> [http://htk.eng.cam.ac.uk/docs/docs.shtml book]<br /> [http://htk.eng.cam.ac.uk/mailing/subscribe_mail.shtml mail-list]<br /> |[http://htk.eng.cam.ac.uk/extensions/index.shtml official], [http://htk.eng.cam.ac.uk/develop/atk.shtml ATK], [https://github.com/ramon-astudillo/custom_fe FE uncertainty decoding]<br /> |C<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMN<br /> |GMM (Full Cov.), Tied-Mix, Streams<br /> |HLDA, MLLR (w/ reg. trees), CMLR (w/ adaptive training), MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch, MMI, MPE, MWE<br /> |{{yes|Yes}}<br /> |[http://catalog.elra.info/product_info.php?cPath=37_40&amp;products_id=693 AURORA2 (purch.)] [http://catalog.elra.info/index.php?cPath=37_40 AURORA3 (purch.)], [http://www.keithv.com/software/htk/ AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/projects/chime/PCC/data/pcchome.tar.gz CHIME-1], [ftp://ftp.dcs.shef.ac.uk/share/spandh/chime_challenge/grid/eval_tools_grid.tgz CHIME-2-I], [http://spandh.dcs.shef.ac.uk/chime_challenge/chime2_task2.html CHIME-2-II],[http://reverb2014.dereverberation.com/tools/REVERB_TOOLS_FOR_ASR_ver2.0.tgz REVERB]<br /> |ETSI-AFE-AURORA2 [http://aurora.hsnr.de/download/Aurora2_afe_v1_1.pdf paper] (see AURORA2 purch.)<br /> |-<br /> !Kaldi<br /> |2009-* (continous updates)<br /> |{{yes|Yes}}<br /> |{{Yes|Apache 2.0}}<br /> |{{some|Windows (not mantained as of 2014), Linux, OSX}}<br /> |[http://kaldi.sourceforge.net/about.html website]<br /> [http://homepages.inf.ed.ac.uk/aghoshal/pubs/asru11-kaldi.pdf paper]<br /> [http://sourceforge.net/p/kaldi/mailman/kaldi-users/ mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://kaldi.sourceforge.net/install.html SVN]<br /> |<br /> |C++<br /> |{{yes|BLAS, LAPACK, GPU (for DNNs)}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMVN<br /> |GMM (Full Cov.), SGMM, DNN<br /> |HLDA, STC, MLLT, MLLR, CMLLR (w/ reg. trees), Exponential transform<br /> |aligment, N-best, lattice rescoring (using OpenFST)<br /> |Baum-Welch, MMI (boosted), MC, feature-based<br /> |{{yes|Yes}}<br /> |[http://kaldi.sourceforge.net/data_prep.html AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/chime_challenge/WSJ0public/CHiME2012-WSJ0-Kaldi_0.03.tar.gz CHIME-2]<br /> |Weniger2014-REVERB [http://reverb2014.dereverberation.com/workshop/reverb2014-papers/1569884459.pdf Paper] [http://www.mmk.ei.tum.de/~wen/REVERB_2014/kaldi_baseline.tar.gz Code]<br /> |-<br /> !Spraak<br /> |2008-* (1.1.374)<br /> |{{yes|Yes}}<br /> |{{no|[http://www.spraak.org/obtaining-spraak/license proprietary]}}<br /> |{{some|Windows ([http://www.spraak.org/obtaining-spraak/system-requirements limited]), Linux, OSX}}<br /> |[http://www.spraak.org/ website]<br /> [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/wambacq/interspeech08/is2008_spraak_v3.pdf paper]<br /> [http://www.spraak.org/mailing-lists mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://www.spraak.org/documentation/doxygen/doc/html/spr__svn.html SVN]<br /> |Missing Data Techniques (MDT)<br /> |C, Python<br /> |{{no|No}} <br /> |{{yes|Yes}}<br /> |Flexible preprocessing script language -- examples for MFCC, PLP <br /> |VTLN,CMN, [http://www.spraak.org/documentation/doxygen/doc/html/spr__tut__mida.html MIDA], [http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__intro.html MDT Techniques], Parametric HistEq [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/ICASSP2010/zhang.pdf&amp;auto&amp;xz:icassp10], Noise normalization [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |GMM (Tied-Mix), Exemplar based [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/dtw/paper_dtw.pdf&amp;auto&amp;kd:icassp11a], NN, CRF, ... (flexible using the preprocessing script) [http://lib.ugent.be/catalog/pug01:4382368]<br /> |CMLLR, eigenvoices, GMM-weight based (NMF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/SPEECOM2013/manuscript_zhang.pdf&amp;auto&amp;xzhang13] -- (all have Matlab dependencies); MAP<br /> |aligment, lattice rescoring, SCRF rescoring (using SCARF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/scarf/paper_scarf.pdf&amp;auto&amp;kd:icassp11b], phone lattice rescoring [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/duchato/eurospeech09/flavor.pdf&amp;auto&amp;jd:intersp09]<br /> |Viterbi<br /> |{{yes|Yes}<br /> |[http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__example.html AURORA4], [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |<br /> |}<br /> <br /> == [[Speaker identification and verification]] ==<br /> <br /> == [[Speech enhancement and separation]] ==<br /> <br /> == [[Other applications]] ==<br /> <br /> == Contribute software ==<br /> To contribute new software, please<br /> * [[Special:UserLogin|create an account]] and login<br /> * go to the wiki page above corresponding to your application; if it does not exist yet, you may create it<br /> * click on the &quot;Edit&quot; link at the top of the page and add a new section for your software (software is ordered by year of the latest version)<br /> * click on the &quot;Save page&quot; link at the bottom of the page to save your modifications<br /> <br /> Please make sure to provide the following information:<br /> * name of the software and year of the latest version<br /> * authors, institution, contact information<br /> * link to the software, ideally including a short demo, and to the external libraries needed<br /> * short description (functionalities, inputs and outputs, programming language, operating system, license, etc) and link to a paper/report describing the software, if any<br /> * whether running on well-known baselines (Aurora-2, Aurora-4, Switchboard, CHiME, etc) is included or requires wrapping by the user<br /> <br /> In order to save storage space, please do not upload the software on this wiki, but link it as much as possible from a public repository (e.g., bitbucket, github, sourceforge) or from a stable URL on the website of your institution. If this is not possible, please contact the [[Main_Page#Working group contacts|resources sharing working group]].</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=Software&diff=312 Software 2014-11-17T18:58:51Z <p>Ramon.astudillo: /* Automatic speech recognition */</p> <hr /> <div>This page provides software grouped by application. <br /> <br /> == [[Automatic speech recognition]] ==<br /> <br /> {| class=&quot;wikitable sortable&quot; style=&quot;font-size:72%; border:gray solid 1px; text-align:center; width:auto; table-layout:fixed;&quot;<br /> |-<br /> !style=&quot;width: 40px&quot; rowspan=&quot;2&quot; class=&quot;unsortable&quot;| ASR engines<br /> !colspan=&quot;6&quot; |General attributes<br /> !colspan=&quot;2&quot; |Programming<br /> !colspan=&quot;8&quot; |Implemented ASR techniques<br /> !colspan=&quot;2&quot; |Reproducible research<br /> |-<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | release / update<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | actively developed<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | licence<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | platforms<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | links <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | extensions<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | language<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | hardware optimization<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | VAD<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic features<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | feature normalization / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic models<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | model adaptation / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | decoding techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | training techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | online ASR<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | robust ASR training recipes <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | reproducible results<br /> |-<br /> !CMU Sphinx<br /> |1986-* (Sphinx 4.1.0, pocketsphinx 0.8) <br /> |{{yes|Yes}}<br /> |{{yes|[https://raw.githubusercontent.com/cmusphinx/sphinx4/master/license.terms BSD-like]}}<br /> |{{yes|Windows, Linux, OSX (Sphinx4) / Raspberry-pi, iPhone, Android (pocketsphinx)}}<br /> |[http://cmusphinx.sourceforge.net/ website] <br /> [http://www.cs.cmu.edu/~rsingh/homepage/papers/icassp03-sphinx4_2.pdf paper]<br /> [https://www.cs.cmu.edu/~awb/papers/ICASSP2006/0100185.pdf paper]<br /> [http://sourceforge.net/p/cmusphinx/mailman/ mail-list]<br /> [http://sourceforge.net/p/cmusphinx/discussion/ forum]<br /> [https://github.com/cmusphinx github]<br /> |<br /> |Java (Sphinx4), C (pocketsphinx)<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |CMN, Mel-Spectrum subtraction<br /> |GMM, Streams<br /> |MLLR, MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch<br /> |{{yes|Yes}}<br /> |[http://www.keithv.com/software/sphinx/ AURORA4 (WSJ0)]<br /> |<br /> |-<br /> !HTK<br /> |1993-2009 (3.4.1) <br /> |{{yes|Yes}}<br /> |{{no|[http://htk.eng.cam.ac.uk/docs/license.shtml proprietary]}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://htk.eng.cam.ac.uk/download.shtml website]<br /> [http://htk.eng.cam.ac.uk/docs/docs.shtml book]<br /> [http://htk.eng.cam.ac.uk/mailing/subscribe_mail.shtml mail-list]<br /> |[http://htk.eng.cam.ac.uk/extensions/index.shtml official], [http://htk.eng.cam.ac.uk/develop/atk.shtml ATK], [https://github.com/ramon-astudillo/custom_fe FE uncertainty decoding]<br /> |C<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMN<br /> |GMM (Full Cov.), Tied-Mix, Streams<br /> |HLDA, MLLR (w/ reg. trees), CMLR (w/ adaptive training), MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch, MMI, MPE, MWE<br /> |{{yes|Yes}}<br /> |[http://catalog.elra.info/product_info.php?cPath=37_40&amp;products_id=693 AURORA2 (purch.)] [http://catalog.elra.info/index.php?cPath=37_40 AURORA3 (purch.)], [http://www.keithv.com/software/htk/ AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/projects/chime/PCC/data/pcchome.tar.gz CHIME-1], [ftp://ftp.dcs.shef.ac.uk/share/spandh/chime_challenge/grid/eval_tools_grid.tgz CHIME-2-I], [http://spandh.dcs.shef.ac.uk/chime_challenge/chime2_task2.html CHIME-2-II],[http://reverb2014.dereverberation.com/tools/REVERB_TOOLS_FOR_ASR_ver2.0.tgz REVERB]<br /> |ETSI-AFE-AURORA2 [http://aurora.hsnr.de/download/Aurora2_afe_v1_1.pdf paper] (see AURORA2 purch.)<br /> |-<br /> !Kaldi<br /> |2009-* (continous updates)<br /> |{{yes|Yes}}<br /> |{{Yes|Apache 2.0}}<br /> |{{some|Windows (not mantained as of 2014), Linux, OSX}}<br /> |[http://kaldi.sourceforge.net/about.html website]<br /> [http://homepages.inf.ed.ac.uk/aghoshal/pubs/asru11-kaldi.pdf paper]<br /> [http://sourceforge.net/p/kaldi/mailman/kaldi-users/ mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://kaldi.sourceforge.net/install.html SVN]<br /> |<br /> |C++<br /> |{{yes|BLAS, LAPACK, GPU (for DNNs)}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMVN<br /> |GMM (Full Cov.), SGMM, DNN<br /> |HLDA, STC, MLLT, MLLR, CMLLR (w/ reg. trees), Exponential transform<br /> |aligment, N-best, lattice rescoring (using OpenFST)<br /> |Baum-Welch, MMI (boosted), MC, feature-based<br /> |{{yes|Yes}}<br /> |[http://kaldi.sourceforge.net/data_prep.html AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/chime_challenge/WSJ0public/CHiME2012-WSJ0-Kaldi_0.03.tar.gz CHIME-2]<br /> |Weniger2014-REVERB [http://reverb2014.dereverberation.com/workshop/reverb2014-papers/1569884459.pdf Paper] [http://www.mmk.ei.tum.de/~wen/REVERB_2014/kaldi_baseline.tar.gz Code]<br /> |-<br /> !Spraak<br /> |2008-* (1.1.374)<br /> |{{yes|Yes}}<br /> |{{no|[http://www.spraak.org/obtaining-spraak/license proprietary]}}<br /> |{{some|Windows ([http://www.spraak.org/obtaining-spraak/system-requirements limited]), Linux, OSX}}<br /> |[http://www.spraak.org/ website]<br /> [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/wambacq/interspeech08/is2008_spraak_v3.pdf paper]<br /> [http://www.spraak.org/mailing-lists mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://www.spraak.org/documentation/doxygen/doc/html/spr__svn.html SVN]<br /> |Missing Data Techniques (MDT)<br /> |C, Python<br /> |{{no|No}} <br /> |{{yes|Yes}}<br /> |Flexible preprocessing script language -- examples for MFCC, PLP <br /> |VTLN,CMN, [http://www.spraak.org/documentation/doxygen/doc/html/spr__tut__mida.html MIDA], [http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__intro.html MDT Techniques], Parametric HistEq [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/ICASSP2010/zhang.pdf&amp;auto&amp;xz:icassp10], Noise normalization [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |GMM (Tied-Mix), Exemplar based [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/dtw/paper_dtw.pdf&amp;auto&amp;kd:icassp11a], NN, CRF, ... (flexible using the preprocessing script) [http://lib.ugent.be/catalog/pug01:4382368]<br /> |CMLLR, eigenvoices, GMM-weight based (NMF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/SPEECOM2013/manuscript_zhang.pdf&amp;auto&amp;xzhang13] -- (all have Matlab dependencies); MAP<br /> |aligment, lattice rescoring, SCRF rescoring (using SCARF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/scarf/paper_scarf.pdf&amp;auto&amp;kd:icassp11b], phone lattice rescoring [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/duchato/eurospeech09/flavor.pdf&amp;auto&amp;jd:intersp09]<br /> |Viterbi<br /> |{{yes|Yes}<br /> |[http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__example.html AURORA4], [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |-<br /> !RWTH<br /> |2001-* (0.6.1)<br /> |{{yes|Yes}}<br /> |{{no|[http://www.spraak.org/obtaining-spraak/license proprietary]}}<br /> |{{some|Windows ([http://www.spraak.org/obtaining-spraak/system-requirements limited]), Linux, OSX}}<br /> |[http://www.spraak.org/ website]<br /> [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/wambacq/interspeech08/is2008_spraak_v3.pdf paper]<br /> [http://www.spraak.org/mailing-lists mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://www.spraak.org/documentation/doxygen/doc/html/spr__svn.html SVN]<br /> |Missing Data Techniques (MDT)<br /> |C, Python<br /> |{{no|No}} <br /> |{{yes|Yes}}<br /> |Flexible preprocessing script language -- examples for MFCC, PLP <br /> |VTLN,CMN, [http://www.spraak.org/documentation/doxygen/doc/html/spr__tut__mida.html MIDA], [http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__intro.html MDT Techniques], Parametric HistEq [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/ICASSP2010/zhang.pdf&amp;auto&amp;xz:icassp10], Noise normalization [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |GMM (Tied-Mix), Exemplar based [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/dtw/paper_dtw.pdf&amp;auto&amp;kd:icassp11a], NN, CRF, ... (flexible using the preprocessing script) [http://lib.ugent.be/catalog/pug01:4382368]<br /> |CMLLR, eigenvoices, GMM-weight based (NMF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/SPEECOM2013/manuscript_zhang.pdf&amp;auto&amp;xzhang13] -- (all have Matlab dependencies); MAP<br /> |aligment, lattice rescoring, SCRF rescoring (using SCARF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/scarf/paper_scarf.pdf&amp;auto&amp;kd:icassp11b], phone lattice rescoring [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/duchato/eurospeech09/flavor.pdf&amp;auto&amp;jd:intersp09]<br /> |Viterbi<br /> |{{yes|Yes}}<br /> |[http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__example.html AURORA4], [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |<br /> |}<br /> <br /> == [[Speaker identification and verification]] ==<br /> <br /> == [[Speech enhancement and separation]] ==<br /> <br /> == [[Other applications]] ==<br /> <br /> == Contribute software ==<br /> To contribute new software, please<br /> * [[Special:UserLogin|create an account]] and login<br /> * go to the wiki page above corresponding to your application; if it does not exist yet, you may create it<br /> * click on the &quot;Edit&quot; link at the top of the page and add a new section for your software (software is ordered by year of the latest version)<br /> * click on the &quot;Save page&quot; link at the bottom of the page to save your modifications<br /> <br /> Please make sure to provide the following information:<br /> * name of the software and year of the latest version<br /> * authors, institution, contact information<br /> * link to the software, ideally including a short demo, and to the external libraries needed<br /> * short description (functionalities, inputs and outputs, programming language, operating system, license, etc) and link to a paper/report describing the software, if any<br /> * whether running on well-known baselines (Aurora-2, Aurora-4, Switchboard, CHiME, etc) is included or requires wrapping by the user<br /> <br /> In order to save storage space, please do not upload the software on this wiki, but link it as much as possible from a public repository (e.g., bitbucket, github, sourceforge) or from a stable URL on the website of your institution. If this is not possible, please contact the [[Main_Page#Working group contacts|resources sharing working group]].</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=Software&diff=311 Software 2014-11-17T18:58:04Z <p>Ramon.astudillo: /* Automatic speech recognition */</p> <hr /> <div>This page provides software grouped by application. <br /> <br /> == [[Automatic speech recognition]] ==<br /> <br /> {| class=&quot;wikitable sortable&quot; style=&quot;font-size:72%; border:gray solid 1px; text-align:center; width:auto; table-layout:fixed;&quot;<br /> |-<br /> !style=&quot;width: 40px&quot; rowspan=&quot;2&quot; class=&quot;unsortable&quot;| ASR engines<br /> !colspan=&quot;6&quot; |General attributes<br /> !colspan=&quot;2&quot; |Programming<br /> !colspan=&quot;8&quot; |Implemented ASR techniques<br /> !colspan=&quot;2&quot; |Reproducible research<br /> |-<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | release / update<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | actively developed<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | licence<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | platforms<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | links <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | extensions<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | language<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | hardware optimization<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | VAD<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic features<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | feature normalization / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic models<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | model adaptation / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | decoding techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | training techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | online ASR<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | robust ASR training recipes <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | reproducible results<br /> |-<br /> !CMU Sphinx<br /> |1986-* (Sphinx 4.1.0, pocketsphinx 0.8) <br /> |{{yes|Yes}}<br /> |{{yes|[https://raw.githubusercontent.com/cmusphinx/sphinx4/master/license.terms BSD-like]}}<br /> |{{yes|Windows, Linux, OSX (Sphinx4) / Raspberry-pi, iPhone, Android (pocketsphinx)}}<br /> |[http://cmusphinx.sourceforge.net/ website] <br /> [http://www.cs.cmu.edu/~rsingh/homepage/papers/icassp03-sphinx4_2.pdf paper]<br /> [https://www.cs.cmu.edu/~awb/papers/ICASSP2006/0100185.pdf paper]<br /> [http://sourceforge.net/p/cmusphinx/mailman/ mail-list]<br /> [http://sourceforge.net/p/cmusphinx/discussion/ forum]<br /> [https://github.com/cmusphinx github]<br /> |<br /> |Java (Sphinx4), C (pocketsphinx)<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |CMN, Mel-Spectrum subtraction<br /> |GMM, Streams<br /> |MLLR, MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch<br /> |{{yes|Yes}}<br /> |[http://www.keithv.com/software/sphinx/ AURORA4 (WSJ0)]<br /> |<br /> |-<br /> !HTK<br /> |1993-2009 (3.4.1) <br /> |{{yes|Yes}}<br /> |{{no|[http://htk.eng.cam.ac.uk/docs/license.shtml proprietary]}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://htk.eng.cam.ac.uk/download.shtml website]<br /> [http://htk.eng.cam.ac.uk/docs/docs.shtml book]<br /> [http://htk.eng.cam.ac.uk/mailing/subscribe_mail.shtml mail-list]<br /> |[http://htk.eng.cam.ac.uk/extensions/index.shtml official], [http://htk.eng.cam.ac.uk/develop/atk.shtml ATK], [https://github.com/ramon-astudillo/custom_fe FE uncertainty decoding]<br /> |C<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMN<br /> |GMM (Full Cov.), Tied-Mix, Streams<br /> |HLDA, MLLR (w/ reg. trees), CMLR (w/ adaptive training), MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch, MMI, MPE, MWE<br /> |{{yes|Yes}}<br /> |[http://catalog.elra.info/product_info.php?cPath=37_40&amp;products_id=693 AURORA2 (purch.)] [http://catalog.elra.info/index.php?cPath=37_40 AURORA3 (purch.)], [http://www.keithv.com/software/htk/ AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/projects/chime/PCC/data/pcchome.tar.gz CHIME-1], [ftp://ftp.dcs.shef.ac.uk/share/spandh/chime_challenge/grid/eval_tools_grid.tgz CHIME-2-I], [http://spandh.dcs.shef.ac.uk/chime_challenge/chime2_task2.html CHIME-2-II],[http://reverb2014.dereverberation.com/tools/REVERB_TOOLS_FOR_ASR_ver2.0.tgz REVERB]<br /> |ETSI-AFE-AURORA2 [http://aurora.hsnr.de/download/Aurora2_afe_v1_1.pdf paper] (see AURORA2 purch.)<br /> |-<br /> !Kaldi<br /> |2009-* (continous updates)<br /> |{{yes|Yes}}<br /> |{{Yes|Apache 2.0}}<br /> |{{some|Windows (not mantained as of 2014), Linux, OSX}}<br /> |[http://kaldi.sourceforge.net/about.html website]<br /> [http://homepages.inf.ed.ac.uk/aghoshal/pubs/asru11-kaldi.pdf paper]<br /> [http://sourceforge.net/p/kaldi/mailman/kaldi-users/ mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://kaldi.sourceforge.net/install.html SVN]<br /> |<br /> |C++<br /> |{{yes|BLAS, LAPACK, GPU (for DNNs)}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMVN<br /> |GMM (Full Cov.), SGMM, DNN<br /> |HLDA, STC, MLLT, MLLR, CMLLR (w/ reg. trees), Exponential transform<br /> |aligment, N-best, lattice rescoring (using OpenFST)<br /> |Baum-Welch, MMI (boosted), MC, feature-based<br /> |{{yes|Yes}}<br /> |[http://kaldi.sourceforge.net/data_prep.html AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/chime_challenge/WSJ0public/CHiME2012-WSJ0-Kaldi_0.03.tar.gz CHIME-2]<br /> |Weniger2014-REVERB [http://reverb2014.dereverberation.com/workshop/reverb2014-papers/1569884459.pdf Paper] [http://www.mmk.ei.tum.de/~wen/REVERB_2014/kaldi_baseline.tar.gz Code]<br /> |-<br /> !Spraak<br /> |2008-* (1.1.374)<br /> |{{yes|Yes}}<br /> |{{no|[http://www.spraak.org/obtaining-spraak/license proprietary]}}<br /> |{{some|Windows ([http://www.spraak.org/obtaining-spraak/system-requirements limited]), Linux, OSX}}<br /> |[http://www.spraak.org/ website]<br /> [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/wambacq/interspeech08/is2008_spraak_v3.pdf paper]<br /> [http://www.spraak.org/mailing-lists mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://www.spraak.org/documentation/doxygen/doc/html/spr__svn.html SVN]<br /> |Missing Data Techniques (MDT)<br /> |C, Python<br /> |{{no|No}} <br /> |{{yes|Yes}}<br /> |Flexible preprocessing script language -- examples for MFCC, PLP <br /> |VTLN,CMN, [http://www.spraak.org/documentation/doxygen/doc/html/spr__tut__mida.html MIDA], [http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__intro.html MDT Techniques], Parametric HistEq [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/ICASSP2010/zhang.pdf&amp;auto&amp;xz:icassp10], Noise normalization [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |GMM (Tied-Mix), Exemplar based [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/dtw/paper_dtw.pdf&amp;auto&amp;kd:icassp11a], NN, CRF, ... (flexible using the preprocessing script) [http://lib.ugent.be/catalog/pug01:4382368]<br /> |CMLLR, eigenvoices, GMM-weight based (NMF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/SPEECOM2013/manuscript_zhang.pdf&amp;auto&amp;xzhang13] -- (all have Matlab dependencies); MAP<br /> |aligment, lattice rescoring, SCRF rescoring (using SCARF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/scarf/paper_scarf.pdf&amp;auto&amp;kd:icassp11b], phone lattice rescoring [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/duchato/eurospeech09/flavor.pdf&amp;auto&amp;jd:intersp09]<br /> |Viterbi<br /> |{{yes|Yes}}<br /> |[http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__example.html AURORA4], [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |-<br /> !Spraak<br /> |2008-* (1.1.374)<br /> |{{yes|Yes}}<br /> |{{no|[http://www.spraak.org/obtaining-spraak/license proprietary]}}<br /> |{{some|Windows ([http://www.spraak.org/obtaining-spraak/system-requirements limited]), Linux, OSX}}<br /> |[http://www.spraak.org/ website]<br /> [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/wambacq/interspeech08/is2008_spraak_v3.pdf paper]<br /> [http://www.spraak.org/mailing-lists mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://www.spraak.org/documentation/doxygen/doc/html/spr__svn.html SVN]<br /> |Missing Data Techniques (MDT)<br /> |C, Python<br /> |{{no|No}} <br /> |{{yes|Yes}}<br /> |Flexible preprocessing script language -- examples for MFCC, PLP <br /> |VTLN,CMN, [http://www.spraak.org/documentation/doxygen/doc/html/spr__tut__mida.html MIDA], [http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__intro.html MDT Techniques], Parametric HistEq [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/ICASSP2010/zhang.pdf&amp;auto&amp;xz:icassp10], Noise normalization [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |GMM (Tied-Mix), Exemplar based [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/dtw/paper_dtw.pdf&amp;auto&amp;kd:icassp11a], NN, CRF, ... (flexible using the preprocessing script) [http://lib.ugent.be/catalog/pug01:4382368]<br /> |CMLLR, eigenvoices, GMM-weight based (NMF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/SPEECOM2013/manuscript_zhang.pdf&amp;auto&amp;xzhang13] -- (all have Matlab dependencies); MAP<br /> |aligment, lattice rescoring, SCRF rescoring (using SCARF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/scarf/paper_scarf.pdf&amp;auto&amp;kd:icassp11b], phone lattice rescoring [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/duchato/eurospeech09/flavor.pdf&amp;auto&amp;jd:intersp09]<br /> |Viterbi<br /> |{{yes|Yes}}<br /> |[http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__example.html AURORA4], [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |<br /> |}<br /> <br /> == [[Speaker identification and verification]] ==<br /> <br /> == [[Speech enhancement and separation]] ==<br /> <br /> == [[Other applications]] ==<br /> <br /> == Contribute software ==<br /> To contribute new software, please<br /> * [[Special:UserLogin|create an account]] and login<br /> * go to the wiki page above corresponding to your application; if it does not exist yet, you may create it<br /> * click on the &quot;Edit&quot; link at the top of the page and add a new section for your software (software is ordered by year of the latest version)<br /> * click on the &quot;Save page&quot; link at the bottom of the page to save your modifications<br /> <br /> Please make sure to provide the following information:<br /> * name of the software and year of the latest version<br /> * authors, institution, contact information<br /> * link to the software, ideally including a short demo, and to the external libraries needed<br /> * short description (functionalities, inputs and outputs, programming language, operating system, license, etc) and link to a paper/report describing the software, if any<br /> * whether running on well-known baselines (Aurora-2, Aurora-4, Switchboard, CHiME, etc) is included or requires wrapping by the user<br /> <br /> In order to save storage space, please do not upload the software on this wiki, but link it as much as possible from a public repository (e.g., bitbucket, github, sourceforge) or from a stable URL on the website of your institution. If this is not possible, please contact the [[Main_Page#Working group contacts|resources sharing working group]].</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=Software&diff=310 Software 2014-11-17T18:56:50Z <p>Ramon.astudillo: /* Automatic speech recognition */</p> <hr /> <div>This page provides software grouped by application. <br /> <br /> == [[Automatic speech recognition]] ==<br /> <br /> {| class=&quot;wikitable sortable&quot; style=&quot;font-size:72%; border:gray solid 1px; text-align:center; width:auto; table-layout:fixed;&quot;<br /> |-<br /> !style=&quot;width: 40px&quot; rowspan=&quot;2&quot; class=&quot;unsortable&quot;| ASR engines<br /> !colspan=&quot;6&quot; |General attributes<br /> !colspan=&quot;2&quot; |Programming<br /> !colspan=&quot;8&quot; |Implemented ASR techniques<br /> !colspan=&quot;2&quot; |Reproducible research<br /> |-<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | release / update<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | actively developed<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | licence<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | platforms<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | links <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | extensions<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | language<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | hardware optimization<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | VAD<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic features<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | feature normalization / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic models<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | model adaptation / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | decoding techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | training techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | online ASR<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | robust ASR training recipes <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | reproducible results<br /> |-<br /> !CMU Sphinx<br /> |1986-* (Sphinx 4.1.0, pocketsphinx 0.8) <br /> |{{yes|Yes}}<br /> |{{yes|[https://raw.githubusercontent.com/cmusphinx/sphinx4/master/license.terms BSD-like]}}<br /> |{{yes|Windows, Linux, OSX (Sphinx4) / Raspberry-pi, iPhone, Android (pocketsphinx)}}<br /> |[http://cmusphinx.sourceforge.net/ website] <br /> [http://www.cs.cmu.edu/~rsingh/homepage/papers/icassp03-sphinx4_2.pdf paper]<br /> [https://www.cs.cmu.edu/~awb/papers/ICASSP2006/0100185.pdf paper]<br /> [http://sourceforge.net/p/cmusphinx/mailman/ mail-list]<br /> [http://sourceforge.net/p/cmusphinx/discussion/ forum]<br /> [https://github.com/cmusphinx github]<br /> |<br /> |Java (Sphinx4), C (pocketsphinx)<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |CMN, Mel-Spectrum subtraction<br /> |GMM, Streams<br /> |MLLR, MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch<br /> |{{yes|Yes}}<br /> |[http://www.keithv.com/software/sphinx/ AURORA4 (WSJ0)]<br /> |<br /> |-<br /> !HTK<br /> |1993-2009 (3.4.1) <br /> |{{yes|Yes}}<br /> |{{no|[http://htk.eng.cam.ac.uk/docs/license.shtml proprietary]}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://htk.eng.cam.ac.uk/download.shtml website]<br /> [http://htk.eng.cam.ac.uk/docs/docs.shtml book]<br /> [http://htk.eng.cam.ac.uk/mailing/subscribe_mail.shtml mail-list]<br /> |[http://htk.eng.cam.ac.uk/extensions/index.shtml official], [http://htk.eng.cam.ac.uk/develop/atk.shtml ATK], [https://github.com/ramon-astudillo/custom_fe FE uncertainty decoding]<br /> |C<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMN<br /> |GMM (Full Cov.), Tied-Mix, Streams<br /> |HLDA, MLLR (w/ reg. trees), CMLR (w/ adaptive training), MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch, MMI, MPE, MWE<br /> |{{yes|Yes}}<br /> |[http://catalog.elra.info/product_info.php?cPath=37_40&amp;products_id=693 AURORA2 (purch.)] [http://catalog.elra.info/index.php?cPath=37_40 AURORA3 (purch.)], [http://www.keithv.com/software/htk/ AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/projects/chime/PCC/data/pcchome.tar.gz CHIME-1], [ftp://ftp.dcs.shef.ac.uk/share/spandh/chime_challenge/grid/eval_tools_grid.tgz CHIME-2-I], [http://spandh.dcs.shef.ac.uk/chime_challenge/chime2_task2.html CHIME-2-II],[http://reverb2014.dereverberation.com/tools/REVERB_TOOLS_FOR_ASR_ver2.0.tgz REVERB]<br /> |ETSI-AFE-AURORA2 [http://aurora.hsnr.de/download/Aurora2_afe_v1_1.pdf paper] (see AURORA2 purch.)<br /> |-<br /> !Kaldi<br /> |2009-* (continous updates)<br /> |{{yes|Yes}}<br /> |{{Yes|Apache 2.0}}<br /> |{{some|Windows (not mantained as of 2014), Linux, OSX}}<br /> |[http://kaldi.sourceforge.net/about.html website]<br /> [http://homepages.inf.ed.ac.uk/aghoshal/pubs/asru11-kaldi.pdf paper]<br /> [http://sourceforge.net/p/kaldi/mailman/kaldi-users/ mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://kaldi.sourceforge.net/install.html SVN]<br /> |<br /> |C++<br /> |{{yes|BLAS, LAPACK, GPU (for DNNs)}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMVN<br /> |GMM (Full Cov.), SGMM, DNN<br /> |HLDA, STC, MLLT, MLLR, CMLLR (w/ reg. trees), Exponential transform<br /> |aligment, N-best, lattice rescoring (using OpenFST)<br /> |Baum-Welch, MMI (boosted), MC, feature-based<br /> |{{yes|Yes}}<br /> |[http://kaldi.sourceforge.net/data_prep.html AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/chime_challenge/WSJ0public/CHiME2012-WSJ0-Kaldi_0.03.tar.gz CHIME-2]<br /> |Weniger2014-REVERB [http://reverb2014.dereverberation.com/workshop/reverb2014-papers/1569884459.pdf Paper] [http://www.mmk.ei.tum.de/~wen/REVERB_2014/kaldi_baseline.tar.gz Code]<br /> |-<br /> !Spraak<br /> |2008-* (1.1.374)<br /> |{{yes|Yes}}<br /> |{{no|[http://www.spraak.org/obtaining-spraak/license proprietary]}}<br /> |{{some|Windows ([http://www.spraak.org/obtaining-spraak/system-requirements limited]), Linux, OSX}}<br /> |[http://www.spraak.org/ website]<br /> [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/wambacq/interspeech08/is2008_spraak_v3.pdf paper]<br /> [http://www.spraak.org/mailing-lists mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://www.spraak.org/documentation/doxygen/doc/html/spr__svn.html SVN]<br /> |Missing Data Techniques (MDT)<br /> |C, Python<br /> |{{no|No}} <br /> |{{yes|Yes}}<br /> |Flexible preprocessing script language -- examples for MFCC, PLP <br /> |VTLN,CMN, [http://www.spraak.org/documentation/doxygen/doc/html/spr__tut__mida.html MIDA], [http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__intro.html MDT Techniques], Parametric HistEq [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/ICASSP2010/zhang.pdf&amp;auto&amp;xz:icassp10], Noise normalization [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |GMM (Tied-Mix), Exemplar based [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/dtw/paper_dtw.pdf&amp;auto&amp;kd:icassp11a], NN, CRF, ... (flexible using the preprocessing script) [http://lib.ugent.be/catalog/pug01:4382368]<br /> |CMLLR, eigenvoices, GMM-weight based (NMF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/SPEECOM2013/manuscript_zhang.pdf&amp;auto&amp;xzhang13] -- (all have Matlab dependencies); MAP<br /> |aligment, lattice rescoring, SCRF rescoring (using SCARF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/scarf/paper_scarf.pdf&amp;auto&amp;kd:icassp11b], phone lattice rescoring [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/duchato/eurospeech09/flavor.pdf&amp;auto&amp;jd:intersp09]<br /> |Viterbi<br /> |{{yes|Yes}}<br /> |[http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__example.html AURORA4], [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |<br /> |}<br /> <br /> == [[Speaker identification and verification]] ==<br /> <br /> == [[Speech enhancement and separation]] ==<br /> <br /> == [[Other applications]] ==<br /> <br /> == Contribute software ==<br /> To contribute new software, please<br /> * [[Special:UserLogin|create an account]] and login<br /> * go to the wiki page above corresponding to your application; if it does not exist yet, you may create it<br /> * click on the &quot;Edit&quot; link at the top of the page and add a new section for your software (software is ordered by year of the latest version)<br /> * click on the &quot;Save page&quot; link at the bottom of the page to save your modifications<br /> <br /> Please make sure to provide the following information:<br /> * name of the software and year of the latest version<br /> * authors, institution, contact information<br /> * link to the software, ideally including a short demo, and to the external libraries needed<br /> * short description (functionalities, inputs and outputs, programming language, operating system, license, etc) and link to a paper/report describing the software, if any<br /> * whether running on well-known baselines (Aurora-2, Aurora-4, Switchboard, CHiME, etc) is included or requires wrapping by the user<br /> <br /> In order to save storage space, please do not upload the software on this wiki, but link it as much as possible from a public repository (e.g., bitbucket, github, sourceforge) or from a stable URL on the website of your institution. If this is not possible, please contact the [[Main_Page#Working group contacts|resources sharing working group]].</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=Software&diff=309 Software 2014-11-17T18:54:27Z <p>Ramon.astudillo: /* Automatic speech recognition */</p> <hr /> <div>This page provides software grouped by application. <br /> <br /> == [[Automatic speech recognition]] ==<br /> <br /> {| class=&quot;wikitable sortable&quot; style=&quot;font-size:72%; border:gray solid 1px; text-align:center; width:auto; table-layout:fixed;&quot;<br /> |-<br /> !style=&quot;width: 40px&quot; rowspan=&quot;2&quot; class=&quot;unsortable&quot;| ASR engines<br /> !colspan=&quot;6&quot; |General attributes<br /> !colspan=&quot;2&quot; |Programming<br /> !colspan=&quot;8&quot; |Implemented ASR techniques<br /> !colspan=&quot;2&quot; |Reproducible research<br /> |-<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | release / update<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | actively developed<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | licence<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | platforms<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | links <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | extensions<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | language<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | hardware optimization<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | VAD<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic features<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | feature normalization / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic models<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | model adaptation / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | decoding techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | training techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | online ASR<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | robust ASR training recipes <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | reproducible results<br /> |-<br /> !CMU Sphinx<br /> |1986-* (Sphinx 4.1.0, pocketsphinx 0.8) <br /> |{{yes|Yes}}<br /> |{{yes|[https://raw.githubusercontent.com/cmusphinx/sphinx4/master/license.terms BSD-like]}}<br /> |{{yes|Windows, Linux, OSX (Sphinx4) / Raspberry-pi, iPhone, Android (pocketsphinx)}}<br /> |[http://cmusphinx.sourceforge.net/ website] <br /> [http://www.cs.cmu.edu/~rsingh/homepage/papers/icassp03-sphinx4_2.pdf paper]<br /> [https://www.cs.cmu.edu/~awb/papers/ICASSP2006/0100185.pdf paper]<br /> [http://sourceforge.net/p/cmusphinx/mailman/ mail-list]<br /> [http://sourceforge.net/p/cmusphinx/discussion/ forum]<br /> [https://github.com/cmusphinx github]<br /> |<br /> |Java (Sphinx4), C (pocketsphinx)<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |CMN, Mel-Spectrum subtraction<br /> |GMM, Streams<br /> |MLLR, MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch<br /> |{{yes|Yes}}<br /> |[http://www.keithv.com/software/sphinx/ AURORA4 (WSJ0)]<br /> |<br /> |-<br /> !HTK<br /> |1993-2009 (3.4.1) <br /> |{{yes|Yes}}<br /> |{{no|[http://htk.eng.cam.ac.uk/docs/license.shtml proprietary]}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://htk.eng.cam.ac.uk/download.shtml website]<br /> [http://htk.eng.cam.ac.uk/docs/docs.shtml book]<br /> [http://htk.eng.cam.ac.uk/mailing/subscribe_mail.shtml mail-list]<br /> |[http://htk.eng.cam.ac.uk/extensions/index.shtml official], [http://htk.eng.cam.ac.uk/develop/atk.shtml ATK], [https://github.com/ramon-astudillo/custom_fe FE uncertainty decoding]<br /> |C<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMN<br /> |GMM (Full Cov.), Tied-Mix, Streams<br /> |HLDA, MLLR (w/ reg. trees), CMLR (w/ adaptive training), MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch, MMI, MPE, MWE<br /> |{{yes|Yes}}<br /> |[http://catalog.elra.info/product_info.php?cPath=37_40&amp;products_id=693 AURORA2 (purch.)] [http://catalog.elra.info/index.php?cPath=37_40 AURORA3 (purch.)], [http://www.keithv.com/software/htk/ AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/projects/chime/PCC/data/pcchome.tar.gz CHIME-1], [ftp://ftp.dcs.shef.ac.uk/share/spandh/chime_challenge/grid/eval_tools_grid.tgz CHIME-2-I], [http://spandh.dcs.shef.ac.uk/chime_challenge/chime2_task2.html CHIME-2-II],[http://reverb2014.dereverberation.com/tools/REVERB_TOOLS_FOR_ASR_ver2.0.tgz REVERB]<br /> |ETSI-AFE-AURORA2 [http://aurora.hsnr.de/download/Aurora2_afe_v1_1.pdf paper] (see AURORA2 purch.)<br /> |-<br /> !Kaldi<br /> |2009-* (continous updates)<br /> |{{yes|Yes}}<br /> |{{Yes|Apache 2.0}}<br /> |{{some|Windows (not mantained as of 2014), Linux, OSX}}<br /> |[http://kaldi.sourceforge.net/about.html website]<br /> [http://homepages.inf.ed.ac.uk/aghoshal/pubs/asru11-kaldi.pdf paper]<br /> [http://sourceforge.net/p/kaldi/mailman/kaldi-users/ mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://kaldi.sourceforge.net/install.html SVN]<br /> |<br /> |C++<br /> |{{yes|BLAS, LAPACK, GPU (for DNNs)}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMVN<br /> |GMM (Full Cov.), SGMM, DNN<br /> |HLDA, STC, MLLT, MLLR, CMLLR (w/ reg. trees), Exponential transform<br /> |aligment, N-best, lattice rescoring (using OpenFST)<br /> |Baum-Welch, MMI (boosted), MC, feature-based<br /> |{{yes|Yes}}<br /> |[http://kaldi.sourceforge.net/data_prep.html AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/chime_challenge/WSJ0public/CHiME2012-WSJ0-Kaldi_0.03.tar.gz CHIME-2]<br /> |Weniger2014-REVERB [http://reverb2014.dereverberation.com/workshop/reverb2014-papers/1569884459.pdf Paper] [http://www.mmk.ei.tum.de/~wen/REVERB_2014/kaldi_baseline.tar.gz Code]<br /> |-<br /> !Spraak<br /> |2008-* (1.1.374)<br /> |{{yes|Yes}}<br /> |{{no|[http://www.spraak.org/obtaining-spraak/license proprietary]}}<br /> |{{some|Windows ([http://www.spraak.org/obtaining-spraak/system-requirements limited]), Linux, OSX}}<br /> |[http://www.spraak.org/ website]<br /> [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/wambacq/interspeech08/is2008_spraak_v3.pdf paper]<br /> [http://www.spraak.org/mailing-lists mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://www.spraak.org/documentation/doxygen/doc/html/spr__svn.html SVN]<br /> |Missing Data Techniques (MDT)<br /> |C, Python<br /> |{{no|No}} <br /> |{{yes|Yes}}<br /> |Flexible preprocessing script language -- examples for MFCC, PLP <br /> |VTLN,CMN, [http://www.spraak.org/documentation/doxygen/doc/html/spr__tut__mida.html MIDA], [http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__intro.html MDT Techniques], Parametric HistEq [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/ICASSP2010/zhang.pdf&amp;auto&amp;xz:icassp10], Noise normalization [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |GMM (Tied-Mix), Exemplar based [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/dtw/paper_dtw.pdf&amp;auto&amp;kd:icassp11a], NN, CRF, ... (flexible using the preprocessing script) [http://lib.ugent.be/catalog/pug01:4382368]<br /> |CMLLR, eigenvoices, GMM-weight based (NMF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/SPEECOM2013/manuscript_zhang.pdf&amp;auto&amp;xzhang13] -- (all have Matlab dependencies); MAP<br /> |aligment, lattice rescoring, SCRF rescoring (using SCARF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/scarf/paper_scarf.pdf&amp;auto&amp;kd:icassp11b], phone lattice rescoring [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/duchato/eurospeech09/flavor.pdf&amp;auto&amp;jd:intersp09]<br /> |Viterbi<br /> |{{yes|Yes}}<br /> |[http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__example.html AURORA4], [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |-<br /> !Julius<br /> |1997-* (4.3.1)<br /> |{{yes|Yes}}<br /> |{{no|proprietary}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |<br /> [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/wambacq/interspeech08/is2008_spraak_v3.pdf paper]<br /> [http://www.spraak.org/mailing-lists mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://www.spraak.org/documentation/doxygen/doc/html/spr__svn.html SVN]<br /> |Missing Data Techniques (MDT)<br /> |C, Python<br /> |{{no|No}} <br /> |{{yes|Yes}}<br /> |Flexible preprocessing script language -- examples for MFCC, PLP <br /> |VTLN,CMN, [http://www.spraak.org/documentation/doxygen/doc/html/spr__tut__mida.html MIDA], [http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__intro.html MDT Techniques], Parametric HistEq [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/ICASSP2010/zhang.pdf&amp;auto&amp;xz:icassp10], Noise normalization [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |GMM (Tied-Mix), Exemplar based [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/dtw/paper_dtw.pdf&amp;auto&amp;kd:icassp11a], NN, CRF, ... (flexible using the preprocessing script) [http://lib.ugent.be/catalog/pug01:4382368]<br /> |CMLLR, eigenvoices, GMM-weight based (NMF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/SPEECOM2013/manuscript_zhang.pdf&amp;auto&amp;xzhang13] -- (all have Matlab dependencies); MAP<br /> |aligment, lattice rescoring, SCRF rescoring (using SCARF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/scarf/paper_scarf.pdf&amp;auto&amp;kd:icassp11b], phone lattice rescoring [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/duchato/eurospeech09/flavor.pdf&amp;auto&amp;jd:intersp09]<br /> |Viterbi<br /> |{{yes|Yes}}<br /> |[http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__example.html AURORA4], [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |<br /> |}<br /> <br /> == [[Speaker identification and verification]] ==<br /> <br /> == [[Speech enhancement and separation]] ==<br /> <br /> == [[Other applications]] ==<br /> <br /> == Contribute software ==<br /> To contribute new software, please<br /> * [[Special:UserLogin|create an account]] and login<br /> * go to the wiki page above corresponding to your application; if it does not exist yet, you may create it<br /> * click on the &quot;Edit&quot; link at the top of the page and add a new section for your software (software is ordered by year of the latest version)<br /> * click on the &quot;Save page&quot; link at the bottom of the page to save your modifications<br /> <br /> Please make sure to provide the following information:<br /> * name of the software and year of the latest version<br /> * authors, institution, contact information<br /> * link to the software, ideally including a short demo, and to the external libraries needed<br /> * short description (functionalities, inputs and outputs, programming language, operating system, license, etc) and link to a paper/report describing the software, if any<br /> * whether running on well-known baselines (Aurora-2, Aurora-4, Switchboard, CHiME, etc) is included or requires wrapping by the user<br /> <br /> In order to save storage space, please do not upload the software on this wiki, but link it as much as possible from a public repository (e.g., bitbucket, github, sourceforge) or from a stable URL on the website of your institution. If this is not possible, please contact the [[Main_Page#Working group contacts|resources sharing working group]].</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=Software&diff=308 Software 2014-11-17T17:48:08Z <p>Ramon.astudillo: /* Automatic speech recognition */</p> <hr /> <div>This page provides software grouped by application. <br /> <br /> == [[Automatic speech recognition]] ==<br /> <br /> {| class=&quot;wikitable sortable&quot; style=&quot;font-size:72%; border:gray solid 1px; text-align:center; width:auto; table-layout:fixed;&quot;<br /> |-<br /> !style=&quot;width: 40px&quot; rowspan=&quot;2&quot; class=&quot;unsortable&quot;| ASR engines<br /> !colspan=&quot;6&quot; |General attributes<br /> !colspan=&quot;2&quot; |Programming<br /> !colspan=&quot;8&quot; |Implemented ASR techniques<br /> !colspan=&quot;2&quot; |Reproducible research<br /> |-<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | release / update<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | actively developed<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | licence<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | platforms<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | links <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | extensions<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | language<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | hardware optimization<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | VAD<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic features<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | feature normalization / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic models<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | model adaptation / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | decoding techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | training techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | online ASR<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | robust ASR training recipes <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | reproducible results<br /> |-<br /> !CMU Sphinx<br /> |1986-* (Sphinx 4.1.0, pocketsphinx 0.8) <br /> |{{yes|Yes}}<br /> |{{yes|[https://raw.githubusercontent.com/cmusphinx/sphinx4/master/license.terms BSD-like]}}<br /> |{{yes|Windows, Linux, OSX (Sphinx4) / Raspberry-pi, iPhone, Android (pocketsphinx)}}<br /> |[http://cmusphinx.sourceforge.net/ website] <br /> [http://www.cs.cmu.edu/~rsingh/homepage/papers/icassp03-sphinx4_2.pdf paper]<br /> [https://www.cs.cmu.edu/~awb/papers/ICASSP2006/0100185.pdf paper]<br /> [http://sourceforge.net/p/cmusphinx/mailman/ mail-list]<br /> [http://sourceforge.net/p/cmusphinx/discussion/ forum]<br /> [https://github.com/cmusphinx github]<br /> |<br /> |Java (Sphinx4), C (pocketsphinx)<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |CMN, Mel-Spectrum subtraction<br /> |GMM, Streams<br /> |MLLR, MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch<br /> |{{yes|Yes}}<br /> |[http://www.keithv.com/software/sphinx/ AURORA4 (WSJ0)]<br /> |<br /> |-<br /> !HTK<br /> |1993-2009 (3.4.1) <br /> |{{yes|Yes}}<br /> |{{no|[http://htk.eng.cam.ac.uk/docs/license.shtml proprietary]}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://htk.eng.cam.ac.uk/download.shtml website]<br /> [http://htk.eng.cam.ac.uk/docs/docs.shtml book]<br /> [http://htk.eng.cam.ac.uk/mailing/subscribe_mail.shtml mail-list]<br /> |[http://htk.eng.cam.ac.uk/extensions/index.shtml official], [http://htk.eng.cam.ac.uk/develop/atk.shtml ATK], [https://github.com/ramon-astudillo/custom_fe FE uncertainty decoding]<br /> |C<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMN<br /> |GMM (Full Cov.), Tied-Mix, Streams<br /> |HLDA, MLLR (w/ reg. trees), CMLR (w/ adaptive training), MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch, MMI, MPE, MWE<br /> |{{yes|Yes}}<br /> |[http://catalog.elra.info/product_info.php?cPath=37_40&amp;products_id=693 AURORA2 (purch.)] [http://catalog.elra.info/index.php?cPath=37_40 AURORA3 (purch.)], [http://www.keithv.com/software/htk/ AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/projects/chime/PCC/data/pcchome.tar.gz CHIME-1], [ftp://ftp.dcs.shef.ac.uk/share/spandh/chime_challenge/grid/eval_tools_grid.tgz CHIME-2-I], [http://spandh.dcs.shef.ac.uk/chime_challenge/chime2_task2.html CHIME-2-II],[http://reverb2014.dereverberation.com/tools/REVERB_TOOLS_FOR_ASR_ver2.0.tgz REVERB]<br /> |ETSI-AFE-AURORA2 [http://aurora.hsnr.de/download/Aurora2_afe_v1_1.pdf paper] (see AURORA2 purch.)<br /> |-<br /> !Kaldi<br /> |2009-* (continous updates)<br /> |{{yes|Yes}}<br /> |{{Yes|Apache 2.0}}<br /> |{{some|Windows (not mantained as of 2014), Linux, OSX}}<br /> |[http://kaldi.sourceforge.net/about.html website]<br /> [http://homepages.inf.ed.ac.uk/aghoshal/pubs/asru11-kaldi.pdf paper]<br /> [http://sourceforge.net/p/kaldi/mailman/kaldi-users/ mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://kaldi.sourceforge.net/install.html SVN]<br /> |<br /> |C++<br /> |{{yes|BLAS, LAPACK, GPU (for DNNs)}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMVN<br /> |GMM (Full Cov.), SGMM, DNN<br /> |HLDA, STC, MLLT, MLLR, CMLLR (w/ reg. trees), Exponential transform<br /> |aligment, N-best, lattice rescoring (using OpenFST)<br /> |Baum-Welch, MMI (boosted), MC, feature-based<br /> |{{yes|Yes}}<br /> |[http://kaldi.sourceforge.net/data_prep.html AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/chime_challenge/WSJ0public/CHiME2012-WSJ0-Kaldi_0.03.tar.gz CHIME-2]<br /> |Weniger2014-REVERB [http://reverb2014.dereverberation.com/workshop/reverb2014-papers/1569884459.pdf Paper] [http://www.mmk.ei.tum.de/~wen/REVERB_2014/kaldi_baseline.tar.gz Code]<br /> |-<br /> !Spraak<br /> |2008-* (1.1.374)<br /> |{{yes|Yes}}<br /> |{{no|[http://www.spraak.org/obtaining-spraak/license proprietary]}}<br /> |{{some|Windows ([http://www.spraak.org/obtaining-spraak/system-requirements limited]), Linux, OSX}}<br /> |[http://www.spraak.org/ website]<br /> [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/wambacq/interspeech08/is2008_spraak_v3.pdf paper]<br /> [http://www.spraak.org/mailing-lists mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://www.spraak.org/documentation/doxygen/doc/html/spr__svn.html SVN]<br /> |Missing Data Techniques (MDT)<br /> |C, Python<br /> |{{no|No}} <br /> |{{yes|Yes}}<br /> |Flexible preprocessing script language -- examples for MFCC, PLP <br /> |VTLN,CMN, [http://www.spraak.org/documentation/doxygen/doc/html/spr__tut__mida.html MIDA], [http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__intro.html MDT Techniques], Parametric HistEq [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/ICASSP2010/zhang.pdf&amp;auto&amp;xz:icassp10], Noise normalization [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |GMM (Tied-Mix), Exemplar based [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/dtw/paper_dtw.pdf&amp;auto&amp;kd:icassp11a], NN, CRF, ... (flexible using the preprocessing script) [http://lib.ugent.be/catalog/pug01:4382368]<br /> |CMLLR, eigenvoices, GMM-weight based (NMF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/SPEECOM2013/manuscript_zhang.pdf&amp;auto&amp;xzhang13] -- (all have Matlab dependencies); MAP<br /> |aligment, lattice rescoring, SCRF rescoring (using SCARF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/scarf/paper_scarf.pdf&amp;auto&amp;kd:icassp11b], phone lattice rescoring [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/duchato/eurospeech09/flavor.pdf&amp;auto&amp;jd:intersp09]<br /> |Viterbi<br /> |{{yes|Yes}}<br /> |[http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__example.html AURORA4], [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |-<br /> !Julius<br /> |1997-* (4.3.1)<br /> |{{yes|Yes}}<br /> |{{no|proprietary}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://www.spraak.org/ website]<br /> [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/wambacq/interspeech08/is2008_spraak_v3.pdf paper]<br /> [http://www.spraak.org/mailing-lists mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://www.spraak.org/documentation/doxygen/doc/html/spr__svn.html SVN]<br /> |Missing Data Techniques (MDT)<br /> |C, Python<br /> |{{no|No}} <br /> |{{yes|Yes}}<br /> |Flexible preprocessing script language -- examples for MFCC, PLP <br /> |VTLN,CMN, [http://www.spraak.org/documentation/doxygen/doc/html/spr__tut__mida.html MIDA], [http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__intro.html MDT Techniques], Parametric HistEq [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/ICASSP2010/zhang.pdf&amp;auto&amp;xz:icassp10], Noise normalization [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |GMM (Tied-Mix), Exemplar based [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/dtw/paper_dtw.pdf&amp;auto&amp;kd:icassp11a], NN, CRF, ... (flexible using the preprocessing script) [http://lib.ugent.be/catalog/pug01:4382368]<br /> |CMLLR, eigenvoices, GMM-weight based (NMF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/SPEECOM2013/manuscript_zhang.pdf&amp;auto&amp;xzhang13] -- (all have Matlab dependencies); MAP<br /> |aligment, lattice rescoring, SCRF rescoring (using SCARF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/scarf/paper_scarf.pdf&amp;auto&amp;kd:icassp11b], phone lattice rescoring [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/duchato/eurospeech09/flavor.pdf&amp;auto&amp;jd:intersp09]<br /> |Viterbi<br /> |{{yes|Yes}}<br /> |[http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__example.html AURORA4], [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |<br /> |}<br /> <br /> == [[Speaker identification and verification]] ==<br /> <br /> == [[Speech enhancement and separation]] ==<br /> <br /> == [[Other applications]] ==<br /> <br /> == Contribute software ==<br /> To contribute new software, please<br /> * [[Special:UserLogin|create an account]] and login<br /> * go to the wiki page above corresponding to your application; if it does not exist yet, you may create it<br /> * click on the &quot;Edit&quot; link at the top of the page and add a new section for your software (software is ordered by year of the latest version)<br /> * click on the &quot;Save page&quot; link at the bottom of the page to save your modifications<br /> <br /> Please make sure to provide the following information:<br /> * name of the software and year of the latest version<br /> * authors, institution, contact information<br /> * link to the software, ideally including a short demo, and to the external libraries needed<br /> * short description (functionalities, inputs and outputs, programming language, operating system, license, etc) and link to a paper/report describing the software, if any<br /> * whether running on well-known baselines (Aurora-2, Aurora-4, Switchboard, CHiME, etc) is included or requires wrapping by the user<br /> <br /> In order to save storage space, please do not upload the software on this wiki, but link it as much as possible from a public repository (e.g., bitbucket, github, sourceforge) or from a stable URL on the website of your institution. If this is not possible, please contact the [[Main_Page#Working group contacts|resources sharing working group]].</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=Software&diff=307 Software 2014-11-17T17:47:32Z <p>Ramon.astudillo: /* Automatic speech recognition */</p> <hr /> <div>This page provides software grouped by application. <br /> <br /> == [[Automatic speech recognition]] ==<br /> <br /> {| class=&quot;wikitable sortable&quot; style=&quot;font-size:72%; border:gray solid 1px; text-align:center; width:auto; table-layout:fixed;&quot;<br /> |-<br /> !style=&quot;width: 40px&quot; rowspan=&quot;2&quot; class=&quot;unsortable&quot;| ASR engines<br /> !colspan=&quot;6&quot; |General attributes<br /> !colspan=&quot;2&quot; |Programming<br /> !colspan=&quot;8&quot; |Implemented ASR techniques<br /> !colspan=&quot;2&quot; |Reproducible research<br /> |-<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | release / update<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | actively developed<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | licence<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | platforms<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | links <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | extensions<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | language<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | hardware optimization<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | VAD<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic features<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | feature normalization / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic models<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | model adaptation / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | decoding techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | training techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | online ASR<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | robust ASR training recipes <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | reproducible results<br /> |-<br /> !CMU Sphinx<br /> |1986-* (Sphinx 4.1.0, pocketsphinx 0.8) <br /> |{{yes|Yes}}<br /> |{{yes|[https://raw.githubusercontent.com/cmusphinx/sphinx4/master/license.terms BSD-like]}}<br /> |{{yes|Windows, Linux, OSX (Sphinx4) / Raspberry-pi, iPhone, Android (pocketsphinx)}}<br /> |[http://cmusphinx.sourceforge.net/ website] <br /> [http://www.cs.cmu.edu/~rsingh/homepage/papers/icassp03-sphinx4_2.pdf paper]<br /> [https://www.cs.cmu.edu/~awb/papers/ICASSP2006/0100185.pdf paper]<br /> [http://sourceforge.net/p/cmusphinx/mailman/ mail-list]<br /> [http://sourceforge.net/p/cmusphinx/discussion/ forum]<br /> [https://github.com/cmusphinx github]<br /> |<br /> |Java (Sphinx4), C (pocketsphinx)<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |CMN, Mel-Spectrum subtraction<br /> |GMM, Streams<br /> |MLLR, MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch<br /> |{{yes|Yes}}<br /> |[http://www.keithv.com/software/sphinx/ AURORA4 (WSJ0)]<br /> |<br /> |-<br /> !HTK<br /> |1993-2009 (3.4.1) <br /> |{{yes|Yes}}<br /> |{{no|[http://htk.eng.cam.ac.uk/docs/license.shtml proprietary]}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://htk.eng.cam.ac.uk/download.shtml website]<br /> [http://htk.eng.cam.ac.uk/docs/docs.shtml book]<br /> [http://htk.eng.cam.ac.uk/mailing/subscribe_mail.shtml mail-list]<br /> |[http://htk.eng.cam.ac.uk/extensions/index.shtml official], [http://htk.eng.cam.ac.uk/develop/atk.shtml ATK], [https://github.com/ramon-astudillo/custom_fe FE uncertainty decoding]<br /> |C<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMN<br /> |GMM (Full Cov.), Tied-Mix, Streams<br /> |HLDA, MLLR (w/ reg. trees), CMLR (w/ adaptive training), MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch, MMI, MPE, MWE<br /> |{{yes|Yes}}<br /> |[http://catalog.elra.info/product_info.php?cPath=37_40&amp;products_id=693 AURORA2 (purch.)] [http://catalog.elra.info/index.php?cPath=37_40 AURORA3 (purch.)], [http://www.keithv.com/software/htk/ AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/projects/chime/PCC/data/pcchome.tar.gz CHIME-1], [ftp://ftp.dcs.shef.ac.uk/share/spandh/chime_challenge/grid/eval_tools_grid.tgz CHIME-2-I], [http://spandh.dcs.shef.ac.uk/chime_challenge/chime2_task2.html CHIME-2-II],[http://reverb2014.dereverberation.com/tools/REVERB_TOOLS_FOR_ASR_ver2.0.tgz REVERB]<br /> |ETSI-AFE-AURORA2 [http://aurora.hsnr.de/download/Aurora2_afe_v1_1.pdf paper] (see AURORA2 purch.)<br /> |-<br /> !Kaldi<br /> |2009-* (continous updates)<br /> |{{yes|Yes}}<br /> |{{Yes|Apache 2.0}}<br /> |{{some|Windows (not mantained as of 2014), Linux, OSX}}<br /> |[http://kaldi.sourceforge.net/about.html website]<br /> [http://homepages.inf.ed.ac.uk/aghoshal/pubs/asru11-kaldi.pdf paper]<br /> [http://sourceforge.net/p/kaldi/mailman/kaldi-users/ mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://kaldi.sourceforge.net/install.html SVN]<br /> |<br /> |C++<br /> |{{yes|BLAS, LAPACK, GPU (for DNNs)}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMVN<br /> |GMM (Full Cov.), SGMM, DNN<br /> |HLDA, STC, MLLT, MLLR, CMLLR (w/ reg. trees), Exponential transform<br /> |aligment, N-best, lattice rescoring (using OpenFST)<br /> |Baum-Welch, MMI (boosted), MC, feature-based<br /> |{{yes|Yes}}<br /> |[http://kaldi.sourceforge.net/data_prep.html AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/chime_challenge/WSJ0public/CHiME2012-WSJ0-Kaldi_0.03.tar.gz CHIME-2]<br /> |Weniger2014-REVERB [http://reverb2014.dereverberation.com/workshop/reverb2014-papers/1569884459.pdf Paper] [http://www.mmk.ei.tum.de/~wen/REVERB_2014/kaldi_baseline.tar.gz Code]<br /> |-<br /> !Spraak<br /> |2008-* (1.1.374)<br /> |{{yes|Yes}}<br /> |{{no|[http://www.spraak.org/obtaining-spraak/license proprietary]}}<br /> |{{some|Windows ([http://www.spraak.org/obtaining-spraak/system-requirements limited]), Linux, OSX}}<br /> |[http://www.spraak.org/ website]<br /> [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/wambacq/interspeech08/is2008_spraak_v3.pdf paper]<br /> [http://www.spraak.org/mailing-lists mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://www.spraak.org/documentation/doxygen/doc/html/spr__svn.html SVN]<br /> |Missing Data Techniques (MDT)<br /> |C, Python<br /> |{{no|No}} <br /> |{{yes|Yes}}<br /> |Flexible preprocessing script language -- examples for MFCC, PLP <br /> |VTLN,CMN, [http://www.spraak.org/documentation/doxygen/doc/html/spr__tut__mida.html MIDA], [http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__intro.html MDT Techniques], Parametric HistEq [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/ICASSP2010/zhang.pdf&amp;auto&amp;xz:icassp10], Noise normalization [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |GMM (Tied-Mix), Exemplar based [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/dtw/paper_dtw.pdf&amp;auto&amp;kd:icassp11a], NN, CRF, ... (flexible using the preprocessing script) [http://lib.ugent.be/catalog/pug01:4382368]<br /> |CMLLR, eigenvoices, GMM-weight based (NMF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/SPEECOM2013/manuscript_zhang.pdf&amp;auto&amp;xzhang13] -- (all have Matlab dependencies); MAP<br /> |aligment, lattice rescoring, SCRF rescoring (using SCARF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/scarf/paper_scarf.pdf&amp;auto&amp;kd:icassp11b], phone lattice rescoring [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/duchato/eurospeech09/flavor.pdf&amp;auto&amp;jd:intersp09]<br /> |Viterbi<br /> |{{yes|Yes}}<br /> |[http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__example.html AURORA4], [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |-<br /> !Julius<br /> |1997-* (4.3.1)<br /> |{{yes|Yes}}<br /> |{{no|proprietary}}<br /> |{{some|Windows ([http://www.spraak.org/obtaining-spraak/system-requirements limited]), Linux, OSX}}<br /> |[http://www.spraak.org/ website]<br /> [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/wambacq/interspeech08/is2008_spraak_v3.pdf paper]<br /> [http://www.spraak.org/mailing-lists mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://www.spraak.org/documentation/doxygen/doc/html/spr__svn.html SVN]<br /> |Missing Data Techniques (MDT)<br /> |C, Python<br /> |{{no|No}} <br /> |{{yes|Yes}}<br /> |Flexible preprocessing script language -- examples for MFCC, PLP <br /> |VTLN,CMN, [http://www.spraak.org/documentation/doxygen/doc/html/spr__tut__mida.html MIDA], [http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__intro.html MDT Techniques], Parametric HistEq [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/ICASSP2010/zhang.pdf&amp;auto&amp;xz:icassp10], Noise normalization [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |GMM (Tied-Mix), Exemplar based [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/dtw/paper_dtw.pdf&amp;auto&amp;kd:icassp11a], NN, CRF, ... (flexible using the preprocessing script) [http://lib.ugent.be/catalog/pug01:4382368]<br /> |CMLLR, eigenvoices, GMM-weight based (NMF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/SPEECOM2013/manuscript_zhang.pdf&amp;auto&amp;xzhang13] -- (all have Matlab dependencies); MAP<br /> |aligment, lattice rescoring, SCRF rescoring (using SCARF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/scarf/paper_scarf.pdf&amp;auto&amp;kd:icassp11b], phone lattice rescoring [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/duchato/eurospeech09/flavor.pdf&amp;auto&amp;jd:intersp09]<br /> |Viterbi<br /> |{{yes|Yes}}<br /> |[http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__example.html AURORA4], [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |<br /> |}<br /> <br /> == [[Speaker identification and verification]] ==<br /> <br /> == [[Speech enhancement and separation]] ==<br /> <br /> == [[Other applications]] ==<br /> <br /> == Contribute software ==<br /> To contribute new software, please<br /> * [[Special:UserLogin|create an account]] and login<br /> * go to the wiki page above corresponding to your application; if it does not exist yet, you may create it<br /> * click on the &quot;Edit&quot; link at the top of the page and add a new section for your software (software is ordered by year of the latest version)<br /> * click on the &quot;Save page&quot; link at the bottom of the page to save your modifications<br /> <br /> Please make sure to provide the following information:<br /> * name of the software and year of the latest version<br /> * authors, institution, contact information<br /> * link to the software, ideally including a short demo, and to the external libraries needed<br /> * short description (functionalities, inputs and outputs, programming language, operating system, license, etc) and link to a paper/report describing the software, if any<br /> * whether running on well-known baselines (Aurora-2, Aurora-4, Switchboard, CHiME, etc) is included or requires wrapping by the user<br /> <br /> In order to save storage space, please do not upload the software on this wiki, but link it as much as possible from a public repository (e.g., bitbucket, github, sourceforge) or from a stable URL on the website of your institution. If this is not possible, please contact the [[Main_Page#Working group contacts|resources sharing working group]].</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=Software&diff=306 Software 2014-11-17T17:45:23Z <p>Ramon.astudillo: /* Automatic speech recognition */</p> <hr /> <div>This page provides software grouped by application. <br /> <br /> == [[Automatic speech recognition]] ==<br /> <br /> {| class=&quot;wikitable sortable&quot; style=&quot;font-size:72%; border:gray solid 1px; text-align:center; width:auto; table-layout:fixed;&quot;<br /> |-<br /> !style=&quot;width: 40px&quot; rowspan=&quot;2&quot; class=&quot;unsortable&quot;| ASR engines<br /> !colspan=&quot;6&quot; |General attributes<br /> !colspan=&quot;2&quot; |Programming<br /> !colspan=&quot;8&quot; |Implemented ASR techniques<br /> !colspan=&quot;2&quot; |Reproducible research<br /> |-<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | release / update<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | actively developed<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | licence<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | platforms<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | links <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | extensions<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | language<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | hardware optimization<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | VAD<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic features<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | feature normalization / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic models<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | model adaptation / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | decoding techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | training techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | online ASR<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | robust ASR training recipes <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | reproducible results<br /> |-<br /> !CMU Sphinx<br /> |1986-* (Sphinx 4.1.0, pocketsphinx 0.8) <br /> |{{yes|Yes}}<br /> |{{yes|[https://raw.githubusercontent.com/cmusphinx/sphinx4/master/license.terms BSD-like]}}<br /> |{{yes|Windows, Linux, OSX (Sphinx4) / Raspberry-pi, iPhone, Android (pocketsphinx)}}<br /> |[http://cmusphinx.sourceforge.net/ website] <br /> [http://www.cs.cmu.edu/~rsingh/homepage/papers/icassp03-sphinx4_2.pdf paper]<br /> [https://www.cs.cmu.edu/~awb/papers/ICASSP2006/0100185.pdf paper]<br /> [http://sourceforge.net/p/cmusphinx/mailman/ mail-list]<br /> [http://sourceforge.net/p/cmusphinx/discussion/ forum]<br /> [https://github.com/cmusphinx github]<br /> |<br /> |Java (Sphinx4), C (pocketsphinx)<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |CMN, Mel-Spectrum subtraction<br /> |GMM, Streams<br /> |MLLR, MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch<br /> |{{yes|Yes}}<br /> |[http://www.keithv.com/software/sphinx/ AURORA4 (WSJ0)]<br /> |<br /> |-<br /> !HTK<br /> |1993-2009 (3.4.1) <br /> |{{yes|Yes}}<br /> |{{no|[http://htk.eng.cam.ac.uk/docs/license.shtml proprietary]}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://htk.eng.cam.ac.uk/download.shtml website]<br /> [http://htk.eng.cam.ac.uk/docs/docs.shtml book]<br /> [http://htk.eng.cam.ac.uk/mailing/subscribe_mail.shtml mail-list]<br /> |[http://htk.eng.cam.ac.uk/extensions/index.shtml official], [http://htk.eng.cam.ac.uk/develop/atk.shtml ATK], [https://github.com/ramon-astudillo/custom_fe FE uncertainty decoding]<br /> |C<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMN<br /> |GMM (Full Cov.), Tied-Mix, Streams<br /> |HLDA, MLLR (w/ reg. trees), CMLR (w/ adaptive training), MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch, MMI, MPE, MWE<br /> |{{yes|Yes}}<br /> |[http://catalog.elra.info/product_info.php?cPath=37_40&amp;products_id=693 AURORA2 (purch.)] [http://catalog.elra.info/index.php?cPath=37_40 AURORA3 (purch.)], [http://www.keithv.com/software/htk/ AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/projects/chime/PCC/data/pcchome.tar.gz CHIME-1], [ftp://ftp.dcs.shef.ac.uk/share/spandh/chime_challenge/grid/eval_tools_grid.tgz CHIME-2-I], [http://spandh.dcs.shef.ac.uk/chime_challenge/chime2_task2.html CHIME-2-II],[http://reverb2014.dereverberation.com/tools/REVERB_TOOLS_FOR_ASR_ver2.0.tgz REVERB]<br /> |ETSI-AFE-AURORA2 [http://aurora.hsnr.de/download/Aurora2_afe_v1_1.pdf paper] (see AURORA2 purch.)<br /> |-<br /> !Kaldi<br /> |2009-* (continous updates)<br /> |{{yes|Yes}}<br /> |{{Yes|Apache 2.0}}<br /> |{{some|Windows (not mantained as of 2014), Linux, OSX}}<br /> |[http://kaldi.sourceforge.net/about.html website]<br /> [http://homepages.inf.ed.ac.uk/aghoshal/pubs/asru11-kaldi.pdf paper]<br /> [http://sourceforge.net/p/kaldi/mailman/kaldi-users/ mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://kaldi.sourceforge.net/install.html SVN]<br /> |<br /> |C++<br /> |{{yes|BLAS, LAPACK, GPU (for DNNs)}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMVN<br /> |GMM (Full Cov.), SGMM, DNN<br /> |HLDA, STC, MLLT, MLLR, CMLLR (w/ reg. trees), Exponential transform<br /> |aligment, N-best, lattice rescoring (using OpenFST)<br /> |Baum-Welch, MMI (boosted), MC, feature-based<br /> |{{yes|Yes}}<br /> |[http://kaldi.sourceforge.net/data_prep.html AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/chime_challenge/WSJ0public/CHiME2012-WSJ0-Kaldi_0.03.tar.gz CHIME-2]<br /> |Weniger2014-REVERB [http://reverb2014.dereverberation.com/workshop/reverb2014-papers/1569884459.pdf Paper] [http://www.mmk.ei.tum.de/~wen/REVERB_2014/kaldi_baseline.tar.gz Code]<br /> |-<br /> !Spraak<br /> |2008-* (1.1.374)<br /> |{{yes|Yes}}<br /> |{{no|[http://www.spraak.org/obtaining-spraak/license proprietary]}}<br /> |{{some|Windows ([http://www.spraak.org/obtaining-spraak/system-requirements limited]), Linux, OSX}}<br /> |[http://www.spraak.org/ website]<br /> [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/wambacq/interspeech08/is2008_spraak_v3.pdf paper]<br /> [http://www.spraak.org/mailing-lists mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://www.spraak.org/documentation/doxygen/doc/html/spr__svn.html SVN]<br /> |Missing Data Techniques (MDT)<br /> |C, Python<br /> |{{no|No}} <br /> |{{yes|Yes}}<br /> |Flexible preprocessing script language -- examples for MFCC, PLP <br /> |VTLN,CMN, [http://www.spraak.org/documentation/doxygen/doc/html/spr__tut__mida.html MIDA], [http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__intro.html MDT Techniques], Parametric HistEq [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/ICASSP2010/zhang.pdf&amp;auto&amp;xz:icassp10], Noise normalization [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |GMM (Tied-Mix), Exemplar based [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/dtw/paper_dtw.pdf&amp;auto&amp;kd:icassp11a], NN, CRF, ... (flexible using the preprocessing script) [http://lib.ugent.be/catalog/pug01:4382368]<br /> |CMLLR, eigenvoices, GMM-weight based (NMF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/SPEECOM2013/manuscript_zhang.pdf&amp;auto&amp;xzhang13] -- (all have Matlab dependencies); MAP<br /> |aligment, lattice rescoring, SCRF rescoring (using SCARF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/scarf/paper_scarf.pdf&amp;auto&amp;kd:icassp11b], phone lattice rescoring [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/duchato/eurospeech09/flavor.pdf&amp;auto&amp;jd:intersp09]<br /> |Viterbi<br /> |{{yes|Yes}}<br /> |[http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__example.html AURORA4], [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |-<br /> !Julius<br /> |1997-* (4.3.1)<br /> |{{yes|Yes}}<br /> |{{no|[http://www.spraak.org/obtaining-spraak/license proprietary]}}<br /> |{{some|Windows ([http://www.spraak.org/obtaining-spraak/system-requirements limited]), Linux, OSX}}<br /> |[http://www.spraak.org/ website]<br /> [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/wambacq/interspeech08/is2008_spraak_v3.pdf paper]<br /> [http://www.spraak.org/mailing-lists mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://www.spraak.org/documentation/doxygen/doc/html/spr__svn.html SVN]<br /> |Missing Data Techniques (MDT)<br /> |C, Python<br /> |{{no|No}} <br /> |{{yes|Yes}}<br /> |Flexible preprocessing script language -- examples for MFCC, PLP <br /> |VTLN,CMN, [http://www.spraak.org/documentation/doxygen/doc/html/spr__tut__mida.html MIDA], [http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__intro.html MDT Techniques], Parametric HistEq [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/ICASSP2010/zhang.pdf&amp;auto&amp;xz:icassp10], Noise normalization [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |GMM (Tied-Mix), Exemplar based [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/dtw/paper_dtw.pdf&amp;auto&amp;kd:icassp11a], NN, CRF, ... (flexible using the preprocessing script) [http://lib.ugent.be/catalog/pug01:4382368]<br /> |CMLLR, eigenvoices, GMM-weight based (NMF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/SPEECOM2013/manuscript_zhang.pdf&amp;auto&amp;xzhang13] -- (all have Matlab dependencies); MAP<br /> |aligment, lattice rescoring, SCRF rescoring (using SCARF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/scarf/paper_scarf.pdf&amp;auto&amp;kd:icassp11b], phone lattice rescoring [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/duchato/eurospeech09/flavor.pdf&amp;auto&amp;jd:intersp09]<br /> |Viterbi<br /> |{{yes|Yes}}<br /> |[http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__example.html AURORA4], [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |<br /> |}<br /> <br /> == [[Speaker identification and verification]] ==<br /> <br /> == [[Speech enhancement and separation]] ==<br /> <br /> == [[Other applications]] ==<br /> <br /> == Contribute software ==<br /> To contribute new software, please<br /> * [[Special:UserLogin|create an account]] and login<br /> * go to the wiki page above corresponding to your application; if it does not exist yet, you may create it<br /> * click on the &quot;Edit&quot; link at the top of the page and add a new section for your software (software is ordered by year of the latest version)<br /> * click on the &quot;Save page&quot; link at the bottom of the page to save your modifications<br /> <br /> Please make sure to provide the following information:<br /> * name of the software and year of the latest version<br /> * authors, institution, contact information<br /> * link to the software, ideally including a short demo, and to the external libraries needed<br /> * short description (functionalities, inputs and outputs, programming language, operating system, license, etc) and link to a paper/report describing the software, if any<br /> * whether running on well-known baselines (Aurora-2, Aurora-4, Switchboard, CHiME, etc) is included or requires wrapping by the user<br /> <br /> In order to save storage space, please do not upload the software on this wiki, but link it as much as possible from a public repository (e.g., bitbucket, github, sourceforge) or from a stable URL on the website of your institution. If this is not possible, please contact the [[Main_Page#Working group contacts|resources sharing working group]].</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=Software&diff=305 Software 2014-11-17T17:45:03Z <p>Ramon.astudillo: /* Automatic speech recognition */</p> <hr /> <div>This page provides software grouped by application. <br /> <br /> == [[Automatic speech recognition]] ==<br /> <br /> {| class=&quot;wikitable sortable&quot; style=&quot;font-size:72%; border:gray solid 1px; text-align:center; width:auto; table-layout:fixed;&quot;<br /> |-<br /> !style=&quot;width: 40px&quot; rowspan=&quot;2&quot; class=&quot;unsortable&quot;| ASR engines<br /> !colspan=&quot;6&quot; |General attributes<br /> !colspan=&quot;2&quot; |Programming<br /> !colspan=&quot;8&quot; |Implemented ASR techniques<br /> !colspan=&quot;2&quot; |Reproducible research<br /> |-<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | release / update<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | actively developed<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | licence<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | platforms<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | links <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | extensions<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | language<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | hardware optimization<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | VAD<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic features<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | feature normalization / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic models<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | model adaptation / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | decoding techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | training techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | online ASR<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | robust ASR training recipes <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | reproducible results<br /> |-<br /> !CMU Sphinx<br /> |1986-* (Sphinx 4.1.0, pocketsphinx 0.8) <br /> |{{yes|Yes}}<br /> |{{yes|[https://raw.githubusercontent.com/cmusphinx/sphinx4/master/license.terms BSD-like]}}<br /> |{{yes|Windows, Linux, OSX (Sphinx4) / Raspberry-pi, iPhone, Android (pocketsphinx)}}<br /> |[http://cmusphinx.sourceforge.net/ website] <br /> [http://www.cs.cmu.edu/~rsingh/homepage/papers/icassp03-sphinx4_2.pdf paper]<br /> [https://www.cs.cmu.edu/~awb/papers/ICASSP2006/0100185.pdf paper]<br /> [http://sourceforge.net/p/cmusphinx/mailman/ mail-list]<br /> [http://sourceforge.net/p/cmusphinx/discussion/ forum]<br /> [https://github.com/cmusphinx github]<br /> |<br /> |Java (Sphinx4), C (pocketsphinx)<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |CMN, Mel-Spectrum subtraction<br /> |GMM, Streams<br /> |MLLR, MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch<br /> |{{yes|Yes}}<br /> |[http://www.keithv.com/software/sphinx/ AURORA4 (WSJ0)]<br /> |<br /> |-<br /> !HTK<br /> |1993-2009 (3.4.1) <br /> |{{yes|Yes}}<br /> |{{no|[http://htk.eng.cam.ac.uk/docs/license.shtml proprietary]}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://htk.eng.cam.ac.uk/download.shtml website]<br /> [http://htk.eng.cam.ac.uk/docs/docs.shtml book]<br /> [http://htk.eng.cam.ac.uk/mailing/subscribe_mail.shtml mail-list]<br /> |[http://htk.eng.cam.ac.uk/extensions/index.shtml official], [http://htk.eng.cam.ac.uk/develop/atk.shtml ATK], [https://github.com/ramon-astudillo/custom_fe FE uncertainty decoding]<br /> |C<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMN<br /> |GMM (Full Cov.), Tied-Mix, Streams<br /> |HLDA, MLLR (w/ reg. trees), CMLR (w/ adaptive training), MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch, MMI, MPE, MWE<br /> |{{yes|Yes}}<br /> |[http://catalog.elra.info/product_info.php?cPath=37_40&amp;products_id=693 AURORA2 (purch.)] [http://catalog.elra.info/index.php?cPath=37_40 AURORA3 (purch.)], [http://www.keithv.com/software/htk/ AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/projects/chime/PCC/data/pcchome.tar.gz CHIME-1], [ftp://ftp.dcs.shef.ac.uk/share/spandh/chime_challenge/grid/eval_tools_grid.tgz CHIME-2-I], [http://spandh.dcs.shef.ac.uk/chime_challenge/chime2_task2.html CHIME-2-II],[http://reverb2014.dereverberation.com/tools/REVERB_TOOLS_FOR_ASR_ver2.0.tgz REVERB]<br /> |ETSI-AFE-AURORA2 [http://aurora.hsnr.de/download/Aurora2_afe_v1_1.pdf paper] (see AURORA2 purch.)<br /> |-<br /> !Kaldi<br /> |2009-* (continous updates)<br /> |{{yes|Yes}}<br /> |{{Yes|Apache 2.0}}<br /> |{{some|Windows (not mantained as of 2014), Linux, OSX}}<br /> |[http://kaldi.sourceforge.net/about.html website]<br /> [http://homepages.inf.ed.ac.uk/aghoshal/pubs/asru11-kaldi.pdf paper]<br /> [http://sourceforge.net/p/kaldi/mailman/kaldi-users/ mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://kaldi.sourceforge.net/install.html SVN]<br /> |<br /> |C++<br /> |{{yes|BLAS, LAPACK, GPU (for DNNs)}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMVN<br /> |GMM (Full Cov.), SGMM, DNN<br /> |HLDA, STC, MLLT, MLLR, CMLLR (w/ reg. trees), Exponential transform<br /> |aligment, N-best, lattice rescoring (using OpenFST)<br /> |Baum-Welch, MMI (boosted), MC, feature-based<br /> |{{yes|Yes}}<br /> |[http://kaldi.sourceforge.net/data_prep.html AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/chime_challenge/WSJ0public/CHiME2012-WSJ0-Kaldi_0.03.tar.gz CHIME-2]<br /> |Weniger2014-REVERB [http://reverb2014.dereverberation.com/workshop/reverb2014-papers/1569884459.pdf Paper] [http://www.mmk.ei.tum.de/~wen/REVERB_2014/kaldi_baseline.tar.gz Code]<br /> |-<br /> !Spraak<br /> |2008-* (1.1.374)<br /> |{{yes|Yes}}<br /> |{{no|[http://www.spraak.org/obtaining-spraak/license proprietary]}}<br /> |{{some|Windows ([http://www.spraak.org/obtaining-spraak/system-requirements limited]), Linux, OSX}}<br /> |[http://www.spraak.org/ website]<br /> [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/wambacq/interspeech08/is2008_spraak_v3.pdf paper]<br /> [http://www.spraak.org/mailing-lists mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://www.spraak.org/documentation/doxygen/doc/html/spr__svn.html SVN]<br /> |Missing Data Techniques (MDT)<br /> |C, Python<br /> |{{no|No}} <br /> |{{yes|Yes}}<br /> |Flexible preprocessing script language -- examples for MFCC, PLP <br /> |VTLN,CMN, [http://www.spraak.org/documentation/doxygen/doc/html/spr__tut__mida.html MIDA], [http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__intro.html MDT Techniques], Parametric HistEq [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/ICASSP2010/zhang.pdf&amp;auto&amp;xz:icassp10], Noise normalization [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |GMM (Tied-Mix), Exemplar based [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/dtw/paper_dtw.pdf&amp;auto&amp;kd:icassp11a], NN, CRF, ... (flexible using the preprocessing script) [http://lib.ugent.be/catalog/pug01:4382368]<br /> |CMLLR, eigenvoices, GMM-weight based (NMF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/SPEECOM2013/manuscript_zhang.pdf&amp;auto&amp;xzhang13] -- (all have Matlab dependencies); MAP<br /> |aligment, lattice rescoring, SCRF rescoring (using SCARF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/scarf/paper_scarf.pdf&amp;auto&amp;kd:icassp11b], phone lattice rescoring [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/duchato/eurospeech09/flavor.pdf&amp;auto&amp;jd:intersp09]<br /> |Viterbi<br /> |{{yes|Yes}}<br /> |[http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__example.html AURORA4], [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |-<br /> !Julius<br /> |2008-* (1.1.374)<br /> |{{yes|Yes}}<br /> |{{no|[http://www.spraak.org/obtaining-spraak/license proprietary]}}<br /> |{{some|Windows ([http://www.spraak.org/obtaining-spraak/system-requirements limited]), Linux, OSX}}<br /> |[http://www.spraak.org/ website]<br /> [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/wambacq/interspeech08/is2008_spraak_v3.pdf paper]<br /> [http://www.spraak.org/mailing-lists mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://www.spraak.org/documentation/doxygen/doc/html/spr__svn.html SVN]<br /> |Missing Data Techniques (MDT)<br /> |C, Python<br /> |{{no|No}} <br /> |{{yes|Yes}}<br /> |Flexible preprocessing script language -- examples for MFCC, PLP <br /> |VTLN,CMN, [http://www.spraak.org/documentation/doxygen/doc/html/spr__tut__mida.html MIDA], [http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__intro.html MDT Techniques], Parametric HistEq [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/ICASSP2010/zhang.pdf&amp;auto&amp;xz:icassp10], Noise normalization [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |GMM (Tied-Mix), Exemplar based [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/dtw/paper_dtw.pdf&amp;auto&amp;kd:icassp11a], NN, CRF, ... (flexible using the preprocessing script) [http://lib.ugent.be/catalog/pug01:4382368]<br /> |CMLLR, eigenvoices, GMM-weight based (NMF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/SPEECOM2013/manuscript_zhang.pdf&amp;auto&amp;xzhang13] -- (all have Matlab dependencies); MAP<br /> |aligment, lattice rescoring, SCRF rescoring (using SCARF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/scarf/paper_scarf.pdf&amp;auto&amp;kd:icassp11b], phone lattice rescoring [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/duchato/eurospeech09/flavor.pdf&amp;auto&amp;jd:intersp09]<br /> |Viterbi<br /> |{{yes|Yes}}<br /> |[http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__example.html AURORA4], [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |<br /> |}<br /> <br /> == [[Speaker identification and verification]] ==<br /> <br /> == [[Speech enhancement and separation]] ==<br /> <br /> == [[Other applications]] ==<br /> <br /> == Contribute software ==<br /> To contribute new software, please<br /> * [[Special:UserLogin|create an account]] and login<br /> * go to the wiki page above corresponding to your application; if it does not exist yet, you may create it<br /> * click on the &quot;Edit&quot; link at the top of the page and add a new section for your software (software is ordered by year of the latest version)<br /> * click on the &quot;Save page&quot; link at the bottom of the page to save your modifications<br /> <br /> Please make sure to provide the following information:<br /> * name of the software and year of the latest version<br /> * authors, institution, contact information<br /> * link to the software, ideally including a short demo, and to the external libraries needed<br /> * short description (functionalities, inputs and outputs, programming language, operating system, license, etc) and link to a paper/report describing the software, if any<br /> * whether running on well-known baselines (Aurora-2, Aurora-4, Switchboard, CHiME, etc) is included or requires wrapping by the user<br /> <br /> In order to save storage space, please do not upload the software on this wiki, but link it as much as possible from a public repository (e.g., bitbucket, github, sourceforge) or from a stable URL on the website of your institution. If this is not possible, please contact the [[Main_Page#Working group contacts|resources sharing working group]].</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=Software&diff=304 Software 2014-11-17T17:44:19Z <p>Ramon.astudillo: /* Automatic speech recognition */</p> <hr /> <div>This page provides software grouped by application. <br /> <br /> == [[Automatic speech recognition]] ==<br /> <br /> {| class=&quot;wikitable sortable&quot; style=&quot;font-size:72%; border:gray solid 1px; text-align:center; width:auto; table-layout:fixed;&quot;<br /> |-<br /> !style=&quot;width: 40px&quot; rowspan=&quot;2&quot; class=&quot;unsortable&quot;| ASR engines<br /> !colspan=&quot;6&quot; |General attributes<br /> !colspan=&quot;2&quot; |Programming<br /> !colspan=&quot;8&quot; |Implemented ASR techniques<br /> !colspan=&quot;2&quot; |Reproducible research<br /> |-<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | release / update<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | actively developed<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | licence<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | platforms<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | links <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | extensions<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | language<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | hardware optimization<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | VAD<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic features<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | feature normalization / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic models<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | model adaptation / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | decoding techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | training techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | online ASR<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | robust ASR training recipes <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | reproducible results<br /> |-<br /> !CMU Sphinx<br /> |1986-* (Sphinx 4.1.0, pocketsphinx 0.8) <br /> |{{yes|Yes}}<br /> |{{yes|[https://raw.githubusercontent.com/cmusphinx/sphinx4/master/license.terms BSD-like]}}<br /> |{{yes|Windows, Linux, OSX (Sphinx4) / Raspberry-pi, iPhone, Android (pocketsphinx)}}<br /> |[http://cmusphinx.sourceforge.net/ website] <br /> [http://www.cs.cmu.edu/~rsingh/homepage/papers/icassp03-sphinx4_2.pdf paper]<br /> [https://www.cs.cmu.edu/~awb/papers/ICASSP2006/0100185.pdf paper]<br /> [http://sourceforge.net/p/cmusphinx/mailman/ mail-list]<br /> [http://sourceforge.net/p/cmusphinx/discussion/ forum]<br /> [https://github.com/cmusphinx github]<br /> |<br /> |Java (Sphinx4), C (pocketsphinx)<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |CMN, Mel-Spectrum subtraction<br /> |GMM, Streams<br /> |MLLR, MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch<br /> |{{yes|Yes}}<br /> |[http://www.keithv.com/software/sphinx/ AURORA4 (WSJ0)]<br /> |<br /> |-<br /> !HTK<br /> |1993-2009 (3.4.1) <br /> |{{yes|Yes}}<br /> |{{no|[http://htk.eng.cam.ac.uk/docs/license.shtml proprietary]}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://htk.eng.cam.ac.uk/download.shtml website]<br /> [http://htk.eng.cam.ac.uk/docs/docs.shtml book]<br /> [http://htk.eng.cam.ac.uk/mailing/subscribe_mail.shtml mail-list]<br /> |[http://htk.eng.cam.ac.uk/extensions/index.shtml official], [http://htk.eng.cam.ac.uk/develop/atk.shtml ATK], [https://github.com/ramon-astudillo/custom_fe FE uncertainty decoding]<br /> |C<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMN<br /> |GMM (Full Cov.), Tied-Mix, Streams<br /> |HLDA, MLLR (w/ reg. trees), CMLR (w/ adaptive training), MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch, MMI, MPE, MWE<br /> |{{yes|Yes}}<br /> |[http://catalog.elra.info/product_info.php?cPath=37_40&amp;products_id=693 AURORA2 (purch.)] [http://catalog.elra.info/index.php?cPath=37_40 AURORA3 (purch.)], [http://www.keithv.com/software/htk/ AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/projects/chime/PCC/data/pcchome.tar.gz CHIME-1], [ftp://ftp.dcs.shef.ac.uk/share/spandh/chime_challenge/grid/eval_tools_grid.tgz CHIME-2-I], [http://spandh.dcs.shef.ac.uk/chime_challenge/chime2_task2.html CHIME-2-II],[http://reverb2014.dereverberation.com/tools/REVERB_TOOLS_FOR_ASR_ver2.0.tgz REVERB]<br /> |ETSI-AFE-AURORA2 [http://aurora.hsnr.de/download/Aurora2_afe_v1_1.pdf paper] (see AURORA2 purch.)<br /> |-<br /> !Kaldi<br /> |2009-* (continous updates)<br /> |{{yes|Yes}}<br /> |{{Yes|Apache 2.0}}<br /> |{{some|Windows (not mantained as of 2014), Linux, OSX}}<br /> |[http://kaldi.sourceforge.net/about.html website]<br /> [http://homepages.inf.ed.ac.uk/aghoshal/pubs/asru11-kaldi.pdf paper]<br /> [http://sourceforge.net/p/kaldi/mailman/kaldi-users/ mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://kaldi.sourceforge.net/install.html SVN]<br /> |<br /> |C++<br /> |{{yes|BLAS, LAPACK, GPU (for DNNs)}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMVN<br /> |GMM (Full Cov.), SGMM, DNN<br /> |HLDA, STC, MLLT, MLLR, CMLLR (w/ reg. trees), Exponential transform<br /> |aligment, N-best, lattice rescoring (using OpenFST)<br /> |Baum-Welch, MMI (boosted), MC, feature-based<br /> |{{yes|Yes}}<br /> |[http://kaldi.sourceforge.net/data_prep.html AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/chime_challenge/WSJ0public/CHiME2012-WSJ0-Kaldi_0.03.tar.gz CHIME-2]<br /> |Weniger2014-REVERB [http://reverb2014.dereverberation.com/workshop/reverb2014-papers/1569884459.pdf Paper] [http://www.mmk.ei.tum.de/~wen/REVERB_2014/kaldi_baseline.tar.gz Code]<br /> |-<br /> !Spraak<br /> |2008-* (1.1.374)<br /> |{{yes|Yes}}<br /> |{{no|[http://www.spraak.org/obtaining-spraak/license proprietary]}}<br /> |{{some|Windows ([http://www.spraak.org/obtaining-spraak/system-requirements limited]), Linux, OSX}}<br /> |[http://www.spraak.org/ website]<br /> [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/wambacq/interspeech08/is2008_spraak_v3.pdf paper]<br /> [http://www.spraak.org/mailing-lists mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://www.spraak.org/documentation/doxygen/doc/html/spr__svn.html SVN]<br /> |Missing Data Techniques (MDT)<br /> |C, Python<br /> |{{no|No}} <br /> |{{yes|Yes}}<br /> |Flexible preprocessing script language -- examples for MFCC, PLP <br /> |VTLN,CMN, [http://www.spraak.org/documentation/doxygen/doc/html/spr__tut__mida.html MIDA], [http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__intro.html MDT Techniques], Parametric HistEq [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/ICASSP2010/zhang.pdf&amp;auto&amp;xz:icassp10], Noise normalization [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |GMM (Tied-Mix), Exemplar based [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/dtw/paper_dtw.pdf&amp;auto&amp;kd:icassp11a], NN, CRF, ... (flexible using the preprocessing script) [http://lib.ugent.be/catalog/pug01:4382368]<br /> |CMLLR, eigenvoices, GMM-weight based (NMF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/SPEECOM2013/manuscript_zhang.pdf&amp;auto&amp;xzhang13] -- (all have Matlab dependencies); MAP<br /> |aligment, lattice rescoring, SCRF rescoring (using SCARF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/scarf/paper_scarf.pdf&amp;auto&amp;kd:icassp11b], phone lattice rescoring [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/duchato/eurospeech09/flavor.pdf&amp;auto&amp;jd:intersp09]<br /> |Viterbi<br /> |{{yes|Yes}}<br /> |[http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__example.html AURORA4], [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |-<br /> !Spraak<br /> |2008-* (1.1.374)<br /> |{{yes|Yes}}<br /> |{{no|[http://www.spraak.org/obtaining-spraak/license proprietary]}}<br /> |{{some|Windows ([http://www.spraak.org/obtaining-spraak/system-requirements limited]), Linux, OSX}}<br /> |[http://www.spraak.org/ website]<br /> [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/wambacq/interspeech08/is2008_spraak_v3.pdf paper]<br /> [http://www.spraak.org/mailing-lists mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://www.spraak.org/documentation/doxygen/doc/html/spr__svn.html SVN]<br /> |Missing Data Techniques (MDT)<br /> |C, Python<br /> |{{no|No}} <br /> |{{yes|Yes}}<br /> |Flexible preprocessing script language -- examples for MFCC, PLP <br /> |VTLN,CMN, [http://www.spraak.org/documentation/doxygen/doc/html/spr__tut__mida.html MIDA], [http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__intro.html MDT Techniques], Parametric HistEq [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/ICASSP2010/zhang.pdf&amp;auto&amp;xz:icassp10], Noise normalization [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |GMM (Tied-Mix), Exemplar based [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/dtw/paper_dtw.pdf&amp;auto&amp;kd:icassp11a], NN, CRF, ... (flexible using the preprocessing script) [http://lib.ugent.be/catalog/pug01:4382368]<br /> |CMLLR, eigenvoices, GMM-weight based (NMF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/SPEECOM2013/manuscript_zhang.pdf&amp;auto&amp;xzhang13] -- (all have Matlab dependencies); MAP<br /> |aligment, lattice rescoring, SCRF rescoring (using SCARF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/scarf/paper_scarf.pdf&amp;auto&amp;kd:icassp11b], phone lattice rescoring [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/duchato/eurospeech09/flavor.pdf&amp;auto&amp;jd:intersp09]<br /> |Viterbi<br /> |{{yes|Yes}}<br /> |[http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__example.html AURORA4], [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |<br /> |}<br /> <br /> == [[Speaker identification and verification]] ==<br /> <br /> == [[Speech enhancement and separation]] ==<br /> <br /> == [[Other applications]] ==<br /> <br /> == Contribute software ==<br /> To contribute new software, please<br /> * [[Special:UserLogin|create an account]] and login<br /> * go to the wiki page above corresponding to your application; if it does not exist yet, you may create it<br /> * click on the &quot;Edit&quot; link at the top of the page and add a new section for your software (software is ordered by year of the latest version)<br /> * click on the &quot;Save page&quot; link at the bottom of the page to save your modifications<br /> <br /> Please make sure to provide the following information:<br /> * name of the software and year of the latest version<br /> * authors, institution, contact information<br /> * link to the software, ideally including a short demo, and to the external libraries needed<br /> * short description (functionalities, inputs and outputs, programming language, operating system, license, etc) and link to a paper/report describing the software, if any<br /> * whether running on well-known baselines (Aurora-2, Aurora-4, Switchboard, CHiME, etc) is included or requires wrapping by the user<br /> <br /> In order to save storage space, please do not upload the software on this wiki, but link it as much as possible from a public repository (e.g., bitbucket, github, sourceforge) or from a stable URL on the website of your institution. If this is not possible, please contact the [[Main_Page#Working group contacts|resources sharing working group]].</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=List_of_SIG_members&diff=303 List of SIG members 2014-11-17T17:37:12Z <p>Ramon.astudillo: </p> <hr /> <div>111 members as of October 24, 2014:<br /> * Søren Holdt Jensen (Univ. Aalborg)<br /> * Børge Lindberg (Univ. Aalborg)<br /> * Zhen-Hua Tan (Univ. Aalborg)<br /> * Kalle Palomäki (Aalto Univ.)<br /> * Steve Beet (Aculab)<br /> * Gautham Mysore (Adobe)<br /> * Björn Hoffmeister (Amazon)<br /> * Dimitris Dimitriadis (AT&amp;T Research)<br /> * Nassos Katsamanis (TU Athens)<br /> * David Pearce (Audience)<br /> * Wei Xue (Institution of Automation, Chinese Academy of Sciences)<br /> * Keith Ponting (Avaya)<br /> * Daniele Giacobello (Beats by Dr. Dre)<br /> * Armin Sehr (Beuth HS)<br /> * Morgan Nelson (Univ. California, Berkeley)<br /> * Pranav Jawale (IIT Bombay)<br /> * Rajbabu Velmurugan (IIT Bombay)<br /> * Mark Gales (Univ. Cambrigde)<br /> * Angel Mario Castro Martínez (Univ. Carl von Ossietzky)<br /> * Bhiksha Raj (Carnegie-Mellon Univ.)<br /> * Rita Singh (Carnegie-Mellon Univ.)<br /> * Richard Stern (Carnegie-Mellon Univ.)<br /> * Climent Nadeu (UP Catalonia)<br /> * Nestor Becerra Yoma (Univ. Chile)<br /> * Fabio Tesser (CNR)<br /> * Jahangir Alam (CRIM)<br /> * Gerhard Taljaard (CSIR)<br /> * John L. Hansen (UT Dallas)<br /> * Taufiq Hasan (UT Dallas)<br /> * Alexey Tarasov (Dublin IT)<br /> * Emanuël Habets (Univ. Erlangen-Nürnberg)<br /> * Walter Kellermann (Univ. Erlangen-Nürnberg)<br /> * Roland Maas (Univ. Erlangen-Nürnberg)<br /> * Georg Stemmer (Univ. Erlangen-Nürnberg)<br /> * Alessio Brutti (FBK)<br /> * Cristina Guerrero Flores (FBK)<br /> * Waheed Mohamed (FBK)<br /> * Brigitte Richardson (Ford Motors)<br /> * Thad Hughes (Google)<br /> * Kevin Wilson (Google)<br /> * José Andrés Gonzalez (Univ. Granada)<br /> * Antonio M. Peinado (Univ. Granada)<br /> * Pejman Mowlaee (Univ. Graz)<br /> * Kwang Myung Jeon (Gwangju IST)<br /> * Martin Heckmann (Honda)<br /> * Afsaneh Asaei (Idiap)<br /> * Alberto Abad (INESC-Id)<br /> * Ramon F. Astudillo (INESC-Id)<br /> * Arseniy Gorin (Inria)<br /> * Emmanuel Vincent (Inria)<br /> * Dung Tran (Inria)<br /> * Yann Salaün (Inria)<br /> * João Felipe Santos (INRS-EMT)<br /> * Rohit Gowtham (Interactive Intelligence)<br /> * Hynek Hermansky (Johns-Hopkins Univ.)<br /> * Rajesh Hegde (IIT Kanpur)<br /> * Lalan Kumar (IIT Kanpur)<br /> * Karan Nathwani (IIT Kanpur)<br /> * Jort Gemmeke (KU Leuven)<br /> * Hugo van Hamme (KU Leuven)<br /> * Jiri Malek (Univ. Liberec)<br /> * Cong-Thanh Do (LIMSI)<br /> * Rajiv Reddy (LumenVox)<br /> * Alfred Mertins (Univ. Lübeck)<br /> * Florian Müller (Univ. Lübeck)<br /> * Ke Chen (Univ. Manchester)<br /> * Richard Rose (McGill Univ.)<br /> * John Hershey (MERL)<br /> * Jonathan Le Roux (MERL)<br /> * Shinji Watanabe (MERL)<br /> * Mike Seltzer (Microsoft)<br /> * Joseph P. Campbell (MIT)<br /> * Jim Glass (MIT)<br /> * Najim Dehak (MIT)<br /> * Björn Schuller (TU Munich)<br /> * Felix Weninger (TU Munich)<br /> * Lou Boves (Univ. Nijmegen)<br /> * Mario Ganzeboom (Univ. Nijmegen)<br /> * Bert Cranen (Univ. Nijmegen)<br /> * Louis ten Bosch (Univ. Nijmegen)<br /> * David van Leeuwen (Univ. Nijmegen)<br /> * Shuguang Wang (Northwestern Polytechnical Univ., Xi'an)<br /> * Marc Delcroix (NTT)<br /> * Miquel Espi (NTT)<br /> * Keisuke Kinoshita (NTT)<br /> * Cyril Guillaumé (NXP Software)<br /> * Wouter Tirry (NXP Software)<br /> * Arun Narayanan (Ohio State Univ.)<br /> * DeLiang Wang (Ohio State Univ.)<br /> * Simon Bozonnet (OnMobile)<br /> * Reinhold Häb-Umbach (Univ. Paderborn)<br /> * Ananya Misra (Princeton Univ.)<br /> * Ricard Marxer (Univ. Pompeu Fabra)<br /> * Ahmed Hussen Abdelaziz (Ruhr Univ., Bochum)<br /> * Dorothea Kolossa (Ruhr Univ., Bochum)<br /> * Rahil Mahdiyan (Saarland Univ.)<br /> * Hakan Erdogan (Sabanci Univ.)<br /> * Emad Grais (Sabanci Univ.)<br /> * Raghavendra Bilgi (Samsung)<br /> * David Trainor (Sentireal)<br /> * Kenichi Kumatani (Spansion)<br /> * Jordan Cohen (Spelamode)<br /> * Leslie Smith (Univ. Stirling)<br /> * Gérard Chollet (Télécom ParisTech)<br /> * Otman Chakkor (Univ. Tetuan)<br /> * Ngoc Q.K. Duong (Technicolor)<br /> * Zhijian Ou (Tsinghua Univ.)<br /> * Antti Hurmalainen (TU Tampere)<br /> * Tuomas Virtanen (TU Tampere)<br /> * Vinay Modi (Voice Pitara Technologies)<br /> * Jidong Tao (Vlingo)<br /> * Raphaël Blouet</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=Software&diff=302 Software 2014-11-17T11:30:30Z <p>Ramon.astudillo: /* Automatic speech recognition */</p> <hr /> <div>This page provides software grouped by application. <br /> <br /> == [[Automatic speech recognition]] ==<br /> <br /> {| class=&quot;wikitable sortable&quot; style=&quot;font-size:72%; border:gray solid 1px; text-align:center; width:auto; table-layout:fixed;&quot;<br /> |-<br /> !style=&quot;width: 40px&quot; rowspan=&quot;2&quot; class=&quot;unsortable&quot;| ASR engines<br /> !colspan=&quot;6&quot; |General attributes<br /> !colspan=&quot;2&quot; |Programming<br /> !colspan=&quot;8&quot; |Implemented ASR techniques<br /> !colspan=&quot;2&quot; |Reproducible research<br /> |-<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | release / update<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | actively developed<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | licence<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | platforms<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | links <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | extensions<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | language<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | hardware optimization<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | VAD<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic features<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | feature normalization / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic models<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | model adaptation / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | decoding techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | training techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | online ASR<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | robust ASR training recipes <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | reproducible results<br /> |-<br /> !CMU Sphinx<br /> |1986-* (Sphinx 4.1.0, pocketsphinx 0.8) <br /> |{{yes|Yes}}<br /> |{{yes|[https://raw.githubusercontent.com/cmusphinx/sphinx4/master/license.terms BSD-like]}}<br /> |{{yes|Windows, Linux, OSX (Sphinx4) / Raspberry-pi, iPhone, Android (pocketsphinx)}}<br /> |[http://cmusphinx.sourceforge.net/ website] <br /> [http://www.cs.cmu.edu/~rsingh/homepage/papers/icassp03-sphinx4_2.pdf paper]<br /> [https://www.cs.cmu.edu/~awb/papers/ICASSP2006/0100185.pdf paper]<br /> [http://sourceforge.net/p/cmusphinx/mailman/ mail-list]<br /> [http://sourceforge.net/p/cmusphinx/discussion/ forum]<br /> [https://github.com/cmusphinx github]<br /> |<br /> |Java (Sphinx4), C (pocketsphinx)<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |CMN, Mel-Spectrum subtraction<br /> |GMM, Streams<br /> |MLLR, MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch<br /> |{{yes|Yes}}<br /> |[http://www.keithv.com/software/sphinx/ AURORA4 (WSJ0)]<br /> |<br /> |-<br /> !HTK<br /> |1993-2009 (3.4.1) <br /> |{{yes|Yes}}<br /> |{{no|[http://htk.eng.cam.ac.uk/docs/license.shtml proprietary]}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://htk.eng.cam.ac.uk/download.shtml website]<br /> [http://htk.eng.cam.ac.uk/docs/docs.shtml book]<br /> [http://htk.eng.cam.ac.uk/mailing/subscribe_mail.shtml mail-list]<br /> |[http://htk.eng.cam.ac.uk/extensions/index.shtml official], [http://htk.eng.cam.ac.uk/develop/atk.shtml ATK], [https://github.com/ramon-astudillo/custom_fe FE uncertainty decoding]<br /> |C<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMN<br /> |GMM (Full Cov.), Tied-Mix, Streams<br /> |HLDA, MLLR (w/ reg. trees), CMLR (w/ adaptive training), MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch, MMI, MPE, MWE<br /> |{{yes|Yes}}<br /> |[http://catalog.elra.info/product_info.php?cPath=37_40&amp;products_id=693 AURORA2 (purch.)] [http://catalog.elra.info/index.php?cPath=37_40 AURORA3 (purch.)], [http://www.keithv.com/software/htk/ AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/projects/chime/PCC/data/pcchome.tar.gz CHIME-1], [ftp://ftp.dcs.shef.ac.uk/share/spandh/chime_challenge/grid/eval_tools_grid.tgz CHIME-2-I], [http://spandh.dcs.shef.ac.uk/chime_challenge/chime2_task2.html CHIME-2-II],[http://reverb2014.dereverberation.com/tools/REVERB_TOOLS_FOR_ASR_ver2.0.tgz REVERB]<br /> |ETSI-AFE-AURORA2 [http://aurora.hsnr.de/download/Aurora2_afe_v1_1.pdf paper] (see AURORA2 purch.)<br /> |-<br /> !Kaldi<br /> |2009-* (continous updates)<br /> |{{yes|Yes}}<br /> |{{Yes|Apache 2.0}}<br /> |{{some|Windows (not mantained as of 2014), Linux, OSX}}<br /> |[http://kaldi.sourceforge.net/about.html website]<br /> [http://homepages.inf.ed.ac.uk/aghoshal/pubs/asru11-kaldi.pdf paper]<br /> [http://sourceforge.net/p/kaldi/mailman/kaldi-users/ mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://kaldi.sourceforge.net/install.html SVN]<br /> |<br /> |C++<br /> |{{yes|BLAS, LAPACK, GPU (for DNNs)}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMVN<br /> |GMM (Full Cov.), SGMM, DNN<br /> |HLDA, STC, MLLT, MLLR, CMLLR (w/ reg. trees), Exponential transform<br /> |aligment, N-best, lattice rescoring (using OpenFST)<br /> |Baum-Welch, MMI (boosted), MC, feature-based<br /> |{{yes|Yes}}<br /> |[http://kaldi.sourceforge.net/data_prep.html AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/chime_challenge/WSJ0public/CHiME2012-WSJ0-Kaldi_0.03.tar.gz CHIME-2]<br /> |Weniger2014-REVERB [http://reverb2014.dereverberation.com/workshop/reverb2014-papers/1569884459.pdf Paper] [http://www.mmk.ei.tum.de/~wen/REVERB_2014/kaldi_baseline.tar.gz Code]<br /> |-<br /> !Spraak<br /> |2008-* (1.1.374)<br /> |{{yes|Yes}}<br /> |{{no|[http://www.spraak.org/obtaining-spraak/license proprietary]}}<br /> |{{some|Windows ([http://www.spraak.org/obtaining-spraak/system-requirements limited]), Linux, OSX}}<br /> |[http://www.spraak.org/ website]<br /> [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/wambacq/interspeech08/is2008_spraak_v3.pdf paper]<br /> [http://www.spraak.org/mailing-lists mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://www.spraak.org/documentation/doxygen/doc/html/spr__svn.html SVN]<br /> |Missing Data Techniques (MDT)<br /> |C, Python<br /> |{{no|No}} <br /> |{{yes|Yes}}<br /> |Flexible preprocessing script language -- examples for MFCC, PLP <br /> |VTLN,CMN, [http://www.spraak.org/documentation/doxygen/doc/html/spr__tut__mida.html MIDA], [http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__intro.html MDT Techniques], Parametric HistEq [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/ICASSP2010/zhang.pdf&amp;auto&amp;xz:icassp10], Noise normalization [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |GMM (Tied-Mix), Exemplar based [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/dtw/paper_dtw.pdf&amp;auto&amp;kd:icassp11a], NN, CRF, ... (flexible using the preprocessing script) [http://lib.ugent.be/catalog/pug01:4382368]<br /> |CMLLR, eigenvoices, GMM-weight based (NMF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/xzhang/SPEECOM2013/manuscript_zhang.pdf&amp;auto&amp;xzhang13] -- (all have Matlab dependencies); MAP<br /> |aligment, lattice rescoring, SCRF rescoring (using SCARF) [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/icassp11/scarf/paper_scarf.pdf&amp;auto&amp;kd:icassp11b], phone lattice rescoring [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/duchato/eurospeech09/flavor.pdf&amp;auto&amp;jd:intersp09]<br /> |Viterbi<br /> |{{yes|Yes}}<br /> |[http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__example.html AURORA4], [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/krisdm/intersp10/mb_vs_fe.pdf&amp;auto&amp;kd:intersp10]<br /> |<br /> |}<br /> <br /> == [[Speaker identification and verification]] ==<br /> <br /> == [[Speech enhancement and separation]] ==<br /> <br /> == [[Other applications]] ==<br /> <br /> == Contribute software ==<br /> To contribute new software, please<br /> * [[Special:UserLogin|create an account]] and login<br /> * go to the wiki page above corresponding to your application; if it does not exist yet, you may create it<br /> * click on the &quot;Edit&quot; link at the top of the page and add a new section for your software (software is ordered by year of the latest version)<br /> * click on the &quot;Save page&quot; link at the bottom of the page to save your modifications<br /> <br /> Please make sure to provide the following information:<br /> * name of the software and year of the latest version<br /> * authors, institution, contact information<br /> * link to the software, ideally including a short demo, and to the external libraries needed<br /> * short description (functionalities, inputs and outputs, programming language, operating system, license, etc) and link to a paper/report describing the software, if any<br /> * whether running on well-known baselines (Aurora-2, Aurora-4, Switchboard, CHiME, etc) is included or requires wrapping by the user<br /> <br /> In order to save storage space, please do not upload the software on this wiki, but link it as much as possible from a public repository (e.g., bitbucket, github, sourceforge) or from a stable URL on the website of your institution. If this is not possible, please contact the [[Main_Page#Working group contacts|resources sharing working group]].</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=List_of_SIG_members&diff=300 List of SIG members 2014-10-24T10:31:02Z <p>Ramon.astudillo: </p> <hr /> <div>110 members as of October 7, 2014:<br /> * Søren Holdt Jensen (Univ. Aalborg)<br /> * Børge Lindberg (Univ. Aalborg)<br /> * Zhen-Hua Tan (Univ. Aalborg)<br /> * Kalle Palomäki (Aalto Univ.)<br /> * Steve Beet (Aculab)<br /> * Gautham Mysore (Adobe)<br /> * Björn Hoffmeister (Amazon)<br /> * Dimitris Dimitriadis (AT&amp;T Research)<br /> * Nassos Katsamanis (TU Athens)<br /> * David Pearce (Audience)<br /> * Wei Xue (Institution of Automation, Chinese Academy of Sciences)<br /> * Keith Ponting (Avaya)<br /> * Daniele Giacobello (Beats by Dr. Dre)<br /> * Armin Sehr (Beuth HS)<br /> * Morgan Nelson (Univ. California, Berkeley)<br /> * Pranav Jawale (IIT Bombay)<br /> * Rajbabu Velmurugan (IIT Bombay)<br /> * Mark Gales (Univ. Cambrigde)<br /> * Angel Mario Castro Martínez (Univ. Carl von Ossietzky)<br /> * Bhiksha Raj (Carnegie-Mellon Univ.)<br /> * Rita Singh (Carnegie-Mellon Univ.)<br /> * Richard Stern (Carnegie-Mellon Univ.)<br /> * Climent Nadeu (UP Catalonia)<br /> * Nestor Becerra Yoma (Univ. Chile)<br /> * Fabio Tesser (CNR)<br /> * Jahangir Alam (CRIM)<br /> * Gerhard Taljaard (CSIR)<br /> * John L. Hansen (UT Dallas)<br /> * Taufiq Hasan (UT Dallas)<br /> * Alexey Tarasov (Dublin IT)<br /> * Emanuël Habets (Univ. Erlangen-Nürnberg)<br /> * Walter Kellermann (Univ. Erlangen-Nürnberg)<br /> * Roland Maas (Univ. Erlangen-Nürnberg)<br /> * Georg Stemmer (Univ. Erlangen-Nürnberg)<br /> * Alessio Brutti (FBK)<br /> * Cristina Guerrero Flores (FBK)<br /> * Waheed Mohamed (FBK)<br /> * Brigitte Richardson (Ford Motors)<br /> * Thad Hughes (Google)<br /> * Kevin Wilson (Google)<br /> * José Andrés Gonzalez (Univ. Granada)<br /> * Antonio M. Peinado (Univ. Granada)<br /> * Pejman Mowlaee (Univ. Graz)<br /> * Kwang Myung Jeon (Gwangju IST)<br /> * Martin Heckmann (Honda)<br /> * Afsaneh Asaei (Idiap)<br /> * Alberto Abad (INESC-Id)<br /> * Ramon F. Astudillo (INESC-Id)<br /> * Arseniy Gorin (Inria)<br /> * Emmanuel Vincent (Inria)<br /> * Dung Tran (Inria)<br /> * Yann Salaün (Inria)<br /> * João Felipe Santos (INRS-EMT)<br /> * Rohit Gowtham (Interactive Intelligence)<br /> * Hynek Hermansky (Johns-Hopkins Univ.)<br /> * Rajesh Hegde (IIT Kanpur)<br /> * Lalan Kumar (IIT Kanpur)<br /> * Karan Nathwani (IIT Kanpur)<br /> * Jort Gemmeke (KU Leuven)<br /> * Hugo van Hamme (KU Leuven)<br /> * Jiri Malek (Univ. Liberec)<br /> * Cong-Thanh Do (LIMSI)<br /> * Rajiv Reddy (LumenVox)<br /> * Alfred Mertins (Univ. Lübeck)<br /> * Florian Müller (Univ. Lübeck)<br /> * Ke Chen (Univ. Manchester)<br /> * Richard Rose (McGill Univ.)<br /> * John Hershey (MERL)<br /> * Jonathan Le Roux (MERL)<br /> * Shinji Watanabe (MERL)<br /> * Mike Seltzer (Microsoft)<br /> * Joseph P. Campbell (MIT)<br /> * Jim Glass (MIT)<br /> * Najim Dehak (MIT)<br /> * Björn Schuller (TU Munich)<br /> * Felix Weninger (TU Munich)<br /> * Lou Boves (Univ. Nijmegen)<br /> * Mario Ganzeboom (Univ. Nijmegen)<br /> * Bert Cranen (Univ. Nijmegen)<br /> * Louis ten Bosch (Univ. Nijmegen)<br /> * David van Leeuwen (Univ. Nijmegen)<br /> * Shuguang Wang (Northwestern Polytechnical Univ., Xi'an)<br /> * Marc Delcroix (NTT)<br /> * Miquel Espi (NTT)<br /> * Keisuke Kinoshita (NTT)<br /> * Cyril Guillaumé (NXP Software)<br /> * Wouter Tirry (NXP Software)<br /> * Arun Narayanan (Ohio State Univ.)<br /> * DeLiang Wang (Ohio State Univ.)<br /> * Simon Bozonnet (OnMobile)<br /> * Reinhold Häb-Umbach (Univ. Paderborn)<br /> * Ananya Misra (Princeton Univ.)<br /> * Ahmed Hussen Abdelaziz (Ruhr Univ., Bochum)<br /> * Dorothea Kolossa (Ruhr Univ., Bochum)<br /> * Rahil Mahdiyan (Saarland Univ.)<br /> * Hakan Erdogan (Sabanci Univ.)<br /> * Emad Grais (Sabanci Univ.)<br /> * Raghavendra Bilgi (Samsung)<br /> * David Trainor (Sentireal)<br /> * Kenichi Kumatani (Spansion)<br /> * Jordan Cohen (Spelamode)<br /> * Leslie Smith (Univ. Stirling)<br /> * Gérard Chollet (Télécom ParisTech)<br /> * Otman Chakkor (Univ. Tetuan)<br /> * Ngoc Q.K. Duong (Technicolor)<br /> * Zhijian Ou (Tsinghua Univ.)<br /> * Antti Hurmalainen (TU Tampere)<br /> * Tuomas Virtanen (TU Tampere)<br /> * Vinay Modi (Voice Pitara Technologies)<br /> * Jidong Tao (Vlingo)<br /> * Raphaël Blouet</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=List_of_SIG_members&diff=283 List of SIG members 2014-10-06T14:09:29Z <p>Ramon.astudillo: </p> <hr /> <div>109 members as of September 28, 2014:<br /> * Søren Holdt Jensen (Univ. Aalborg)<br /> * Børge Lindberg (Univ. Aalborg)<br /> * Zhen-Hua Tan (Univ. Aalborg)<br /> * Kalle Palomäki (Aalto Univ.)<br /> * Steve Beet (Aculab)<br /> * Gautham Mysore (Adobe)<br /> * Björn Hoffmeister (Amazon)<br /> * Dimitris Dimitriadis (AT&amp;T Research)<br /> * Nassos Katsamanis (TU Athens)<br /> * David Pearce (Audience)<br /> * Wei Xue (Institution of Automation, Chinese Academy of Sciences)<br /> * Keith Ponting (Avaya)<br /> * Daniele Giacobello (Beats by Dr. Dre)<br /> * Armin Sehr (Beuth HS)<br /> * Morgan Nelson (Univ. California, Berkeley)<br /> * Pranav Jawale (IIT Bombay)<br /> * Rajbabu Velmurugan (IIT Bombay)<br /> * Mark Gales (Univ. Cambrigde)<br /> * Bhiksha Raj (Carnegie-Mellon Univ.)<br /> * Rita Singh (Carnegie-Mellon Univ.)<br /> * Richard Stern (Carnegie-Mellon Univ.)<br /> * Climent Nadeu (UP Catalonia)<br /> * Nestor Becerra Yoma (Univ. Chile)<br /> * Fabio Tesser (CNR)<br /> * Jahangir Alam (CRIM)<br /> * Gerhard Taljaard (CSIR)<br /> * John L. Hansen (UT Dallas)<br /> * Taufiq Hasan (UT Dallas)<br /> * Alexey Tarasov (Dublin IT)<br /> * Emanuël Habets (Univ. Erlangen-Nürnberg)<br /> * Walter Kellermann (Univ. Erlangen-Nürnberg)<br /> * Roland Maas (Univ. Erlangen-Nürnberg)<br /> * Georg Stemmer (Univ. Erlangen-Nürnberg)<br /> * Alessio Brutti (FBK)<br /> * Cristina Guerrero Flores (FBK)<br /> * Waheed Mohamed (FBK)<br /> * Brigitte Richardson (Ford Motors)<br /> * Thad Hughes (Google)<br /> * Kevin Wilson (Google)<br /> * José Andrés Gonzalez (Univ. Granada)<br /> * Antonio M. Peinado (Univ. Granada)<br /> * Pejman Mowlaee (Univ. Graz)<br /> * Kwang Myung Jeon (Gwangju IST)<br /> * Martin Heckmann (Honda)<br /> * Afsaneh Asaei (Idiap)<br /> * Alberto Abad (INESC-Id)<br /> * Ramon F. Astudillo (INESC-Id)<br /> * Arseniy Gorin (Inria)<br /> * Emmanuel Vincent (Inria)<br /> * Dung Tran (Inria)<br /> * Yann Salaün (Inria)<br /> * João Felipe Santos (INRS-EMT)<br /> * Rohit Gowtham (Interactive Intelligence)<br /> * Hynek Hermansky (Johns-Hopkins Univ.)<br /> * Rajesh Hegde (IIT Kanpur)<br /> * Lalan Kumar (IIT Kanpur)<br /> * Karan Nathwani (IIT Kanpur)<br /> * Jort Gemmeke (KU Leuven)<br /> * Hugo van Hamme (KU Leuven)<br /> * Jiri Malek (Univ. Liberec)<br /> * Cong-Thanh Do (LIMSI)<br /> * Rajiv Reddy (LumenVox)<br /> * Alfred Mertins (Univ. Lübeck)<br /> * Florian Müller (Univ. Lübeck)<br /> * Ke Chen (Univ. Manchester)<br /> * Richard Rose (McGill Univ.)<br /> * John Hershey (MERL)<br /> * Jonathan Le Roux (MERL)<br /> * Shinji Watanabe (MERL)<br /> * Mike Seltzer (Microsoft)<br /> * Joseph P. Campbell (MIT)<br /> * Jim Glass (MIT)<br /> * Najim Dehak (MIT)<br /> * Björn Schuller (TU Munich)<br /> * Felix Weninger (TU Munich)<br /> * Lou Boves (Univ. Nijmegen)<br /> * Mario Ganzeboom (Univ. Nijmegen)<br /> * Bert Cranen (Univ. Nijmegen)<br /> * Louis ten Bosch (Univ. Nijmegen)<br /> * David van Leeuwen (Univ. Nijmegen)<br /> * Shuguang Wang (Northwestern Polytechnical Univ., Xi'an)<br /> * Marc Delcroix (NTT)<br /> * Miquel Espi (NTT)<br /> * Keisuke Kinoshita (NTT)<br /> * Cyril Guillaumé (NXP Software)<br /> * Wouter Tirry (NXP Software)<br /> * Arun Narayanan (Ohio State Univ.)<br /> * DeLiang Wang (Ohio State Univ.)<br /> * Simon Bozonnet (OnMobile)<br /> * Reinhold Häb-Umbach (Univ. Paderborn)<br /> * Ananya Misra (Princeton Univ.)<br /> * Ahmed Hussen Abdelaziz (Ruhr Univ., Bochum)<br /> * Dorothea Kolossa (Ruhr Univ., Bochum)<br /> * Rahil Mahdiyan (Saarland Univ.)<br /> * Hakan Erdogan (Sabanci Univ.)<br /> * Emad Grais (Sabanci Univ.)<br /> * Raghavendra Bilgi (Samsung)<br /> * David Trainor (Sentireal)<br /> * Kenichi Kumatani (Spansion)<br /> * Jordan Cohen (Spelamode)<br /> * Leslie Smith (Univ. Stirling)<br /> * Gérard Chollet (Télécom ParisTech)<br /> * Otman Chakkor (Univ. Tetuan)<br /> * Ngoc Q.K. Duong (Technicolor)<br /> * Zhijian Ou (Tsinghua Univ.)<br /> * Antti Hurmalainen (TU Tampere)<br /> * Tuomas Virtanen (TU Tampere)<br /> * Vinay Modi (Voice Pitara Technologies)<br /> * Jidong Tao (Vlingo)<br /> * Raphaël Blouet</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=Software&diff=266 Software 2014-09-08T13:35:33Z <p>Ramon.astudillo: /* Automatic speech recognition */</p> <hr /> <div>This page provides software grouped by application. <br /> <br /> == [[Automatic speech recognition]] ==<br /> <br /> {| class=&quot;wikitable sortable&quot; style=&quot;font-size:72%; border:gray solid 1px; text-align:center; width:auto; table-layout:fixed;&quot;<br /> |-<br /> !style=&quot;width: 40px&quot; rowspan=&quot;2&quot; class=&quot;unsortable&quot;| ASR engines<br /> !colspan=&quot;6&quot; |General attributes<br /> !colspan=&quot;2&quot; |Programming<br /> !colspan=&quot;8&quot; |Implemented ASR techniques<br /> !colspan=&quot;2&quot; |Reproducible research<br /> |-<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | release / update<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | actively developed<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | licence<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | platforms<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | links <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | extensions<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | language<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | hardware optimization<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | VAD<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic features<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | feature normalization / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic models<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | model adaptation / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | decoding techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | training techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | online ASR<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | robust ASR training recipes <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | reproducible results<br /> |-<br /> !CMU Sphinx<br /> |1986-* (Sphinx 4.1.0, pocketsphinx 0.8) <br /> |{{yes|Yes}}<br /> |{{some|[https://raw.githubusercontent.com/cmusphinx/sphinx4/master/license.terms proprietary], allows modif.}}<br /> |{{yes|Windows, Linux, OSX (Sphinx4) / Raspberry-pi, iPhone, Android (pocketsphinx)}}<br /> |[http://cmusphinx.sourceforge.net/ website] <br /> [http://www.cs.cmu.edu/~rsingh/homepage/papers/icassp03-sphinx4_2.pdf paper]<br /> [https://www.cs.cmu.edu/~awb/papers/ICASSP2006/0100185.pdf paper]<br /> [http://sourceforge.net/p/cmusphinx/mailman/ mail-list]<br /> [http://sourceforge.net/p/cmusphinx/discussion/ forum]<br /> [https://github.com/cmusphinx github]<br /> |<br /> |Java (Sphinx4), C (pocketsphinx)<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |CMN, Mel-Spectrum subtraction<br /> |GMM, Streams<br /> |MLLR, MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch<br /> |{{yes|Yes}}<br /> |[http://www.keithv.com/software/sphinx/ AURORA4 (WSJ0)]<br /> |<br /> |-<br /> !HTK<br /> |1993-2009 (3.4.1) <br /> |{{no|No}}<br /> |{{no|[http://htk.eng.cam.ac.uk/docs/license.shtml proprietary]}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://htk.eng.cam.ac.uk/download.shtml website]<br /> [http://htk.eng.cam.ac.uk/docs/docs.shtml book]<br /> [http://htk.eng.cam.ac.uk/mailing/subscribe_mail.shtml mail-list]<br /> |[http://htk.eng.cam.ac.uk/extensions/index.shtml official], [http://htk.eng.cam.ac.uk/develop/atk.shtml ATK], [https://github.com/ramon-astudillo/custom_fe FE uncertainty decoding]<br /> |C<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMN<br /> |GMM (Full Cov.), Tied-Mix, Streams<br /> |HLDA, MLLR (w/ reg. trees), CMLR (w/ adaptive training), MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch, MMI, MPE, MWE<br /> |{{yes|Yes}}<br /> |[http://catalog.elra.info/product_info.php?cPath=37_40&amp;products_id=693 AURORA2 (purch.)] [http://catalog.elra.info/index.php?cPath=37_40 AURORA3 (purch.)], [http://www.keithv.com/software/htk/ AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/projects/chime/PCC/data/pcchome.tar.gz CHIME-1], [ftp://ftp.dcs.shef.ac.uk/share/spandh/chime_challenge/grid/eval_tools_grid.tgz CHIME-2-I], [http://spandh.dcs.shef.ac.uk/chime_challenge/chime2_task2.html CHIME-2-II],[http://reverb2014.dereverberation.com/tools/REVERB_TOOLS_FOR_ASR_ver2.0.tgz REVERB]<br /> |ETSI-AFE-AURORA2 [http://aurora.hsnr.de/download/Aurora2_afe_v1_1.pdf paper] (see AURORA2 purch.)<br /> |-<br /> !Kaldi<br /> |2009-* (continous updates)<br /> |{{yes|Yes}}<br /> |{{Yes|Apache 2.0}}<br /> |{{some|Windows (not mantained as of 2014), Linux, OSX}}<br /> |[http://kaldi.sourceforge.net/about.html website]<br /> [http://homepages.inf.ed.ac.uk/aghoshal/pubs/asru11-kaldi.pdf paper]<br /> [http://sourceforge.net/p/kaldi/mailman/kaldi-users/ mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://kaldi.sourceforge.net/install.html SVN]<br /> |<br /> |C++<br /> |{{yes|BLAS, LAPACK, GPU (for DNNs)}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMVN<br /> |GMM (Full Cov.), SGMM, DNN<br /> |HLDA, STC, MLLT, MLLR, CMLLR (w/ reg. trees), Exponential transform<br /> |aligment, N-best, lattice rescoring (uses OpenFST)<br /> |Baum-Welch, MMI (boosted), MC, feature-based<br /> |{{yes|Yes}}<br /> |[http://kaldi.sourceforge.net/data_prep.html AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/chime_challenge/WSJ0public/CHiME2012-WSJ0-Kaldi_0.03.tar.gz CHIME-2]<br /> |Weniger2014-REVERB [http://reverb2014.dereverberation.com/workshop/reverb2014-papers/1569884459.pdf Paper] [http://www.mmk.ei.tum.de/~wen/REVERB_2014/kaldi_baseline.tar.gz Code]<br /> |-<br /> !Spraak<br /> |2012 (1.1)<br /> |{{no|No}}<br /> |{{no|[http://www.spraak.org/obtaining-spraak/license proprietary]}}<br /> |{{some|Windows ([http://www.spraak.org/obtaining-spraak/system-requirements limited]), Linux}}<br /> |[http://www.spraak.org/ website]<br /> [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/wambacq/interspeech08/is2008_spraak_v3.pdf paper]<br /> [http://www.spraak.org/mailing-lists mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://www.spraak.org/documentation/doxygen/doc/html/spr__svn.html SVN]<br /> |<br /> |C, Python<br /> |{{no|No}} <br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN,CMN, [http://www.spraak.org/documentation/doxygen/doc/html/spr__tut__mida.html MIDA], [http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__intro.html MDT Techniques]<br /> |GMM, Tied-Mix, Exemplar based<br /> |CMLLR<br /> |aligment, N-best, lattice rescoring, parallel latices<br /> |Baum-Welch<br /> |{{dunno}}<br /> |[http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__example.html AURORA4]<br /> |<br /> |}<br /> <br /> == [[Speaker identification and verification]] ==<br /> <br /> == [[Speech enhancement and separation]] ==<br /> <br /> == [[Other applications]] ==<br /> <br /> == Contribute software ==<br /> To contribute new software, please<br /> * [[Special:UserLogin|create an account]] and login<br /> * go to the wiki page above corresponding to your application; if it does not exist yet, you may create it<br /> * click on the &quot;Edit&quot; link at the top of the page and add a new section for your software (software is ordered by year of the latest version)<br /> * click on the &quot;Save page&quot; link at the bottom of the page to save your modifications<br /> <br /> Please make sure to provide the following information:<br /> * name of the software and year of the latest version<br /> * authors, institution, contact information<br /> * link to the software, ideally including a short demo, and to the external libraries needed<br /> * short description (functionalities, inputs and outputs, programming language, operating system, license, etc) and link to a paper/report describing the software, if any<br /> * whether running on well-known baselines (Aurora-2, Aurora-4, Switchboard, CHiME, etc) is included or requires wrapping by the user<br /> <br /> In order to save storage space, please do not upload the software on this wiki, but link it as much as possible from a public repository (e.g., bitbucket, github, sourceforge) or from a stable URL on the website of your institution. If this is not possible, please contact the [[Main_Page#Working group contacts|resources sharing working group]].</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=Software&diff=265 Software 2014-09-08T13:16:52Z <p>Ramon.astudillo: /* Automatic speech recognition */</p> <hr /> <div>This page provides software grouped by application. <br /> <br /> == [[Automatic speech recognition]] ==<br /> <br /> {| class=&quot;wikitable sortable&quot; style=&quot;font-size:72%; border:gray solid 1px; text-align:center; width:auto; table-layout:fixed;&quot;<br /> |-<br /> !style=&quot;width: 40px&quot; rowspan=&quot;2&quot; class=&quot;unsortable&quot;| ASR engines<br /> !colspan=&quot;6&quot; |General attributes<br /> !colspan=&quot;2&quot; |Programming<br /> !colspan=&quot;8&quot; |Implemented ASR techniques<br /> !colspan=&quot;2&quot; |Reproducible research<br /> |-<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | release / update<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | actively developed<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | licence<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | platforms<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | links <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | extensions<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | language<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | hardware optimization<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | VAD<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic features<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | feature normalization / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic models<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | model adaptation / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | decoding techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | training techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | online ASR<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | robust ASR training recipes <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | reproducible results<br /> |-<br /> !CMU Sphinx<br /> |1986-* (Sphinx 4.1.0, pocketsphinx 0.8) <br /> |{{yes|Yes}}<br /> |{{some|[https://raw.githubusercontent.com/cmusphinx/sphinx4/master/license.terms proprietary], allows modif.}}<br /> |{{yes|Windows, Linux, OSX (Sphinx4) / Raspberry-pi, iPhone, Android (pocketsphinx)}}<br /> |[http://cmusphinx.sourceforge.net/ website] <br /> [http://www.cs.cmu.edu/~rsingh/homepage/papers/icassp03-sphinx4_2.pdf paper]<br /> [https://www.cs.cmu.edu/~awb/papers/ICASSP2006/0100185.pdf paper]<br /> [http://sourceforge.net/p/cmusphinx/mailman/ mail-list]<br /> [http://sourceforge.net/p/cmusphinx/discussion/ forum]<br /> [https://github.com/cmusphinx github]<br /> |<br /> |Java (Sphinx4), C (pocketsphinx)<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |CMN, Mel-Spectrum subtraction<br /> |GMM, Streams<br /> |MLLR, MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch<br /> |{{yes|Yes}}<br /> |[http://www.keithv.com/software/sphinx/ AURORA4 (WSJ0)]<br /> |<br /> |-<br /> !HTK<br /> |1993-2009 (3.4.1) <br /> |{{no|No}}<br /> |{{no|[http://htk.eng.cam.ac.uk/docs/license.shtml proprietary]}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://htk.eng.cam.ac.uk/download.shtml website]<br /> [http://htk.eng.cam.ac.uk/docs/docs.shtml book]<br /> [http://htk.eng.cam.ac.uk/mailing/subscribe_mail.shtml mail-list]<br /> |[http://htk.eng.cam.ac.uk/extensions/index.shtml official], [http://htk.eng.cam.ac.uk/develop/atk.shtml ATK], [https://github.com/ramon-astudillo/custom_fe uncertainty decoding]<br /> |C<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMN<br /> |GMM (Full Cov.), Tied-Mix, Streams<br /> |HLDA, MLLR (w/ reg. trees), CMLR (w/ adaptive training), MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch, MMI, MPE, MWE<br /> |{{yes|Yes}}<br /> |[http://catalog.elra.info/product_info.php?cPath=37_40&amp;products_id=693 AURORA2 (purch.)] [http://catalog.elra.info/index.php?cPath=37_40 AURORA3 (purch.)], [http://www.keithv.com/software/htk/ AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/projects/chime/PCC/data/pcchome.tar.gz CHIME-1], [ftp://ftp.dcs.shef.ac.uk/share/spandh/chime_challenge/grid/eval_tools_grid.tgz CHIME-2-I], [http://spandh.dcs.shef.ac.uk/chime_challenge/chime2_task2.html CHIME-2-II],[http://reverb2014.dereverberation.com/tools/REVERB_TOOLS_FOR_ASR_ver2.0.tgz REVERB]<br /> |ETSI-AFE-AURORA2 [http://aurora.hsnr.de/download/Aurora2_afe_v1_1.pdf paper] (see AURORA2 purch.)<br /> |-<br /> !Kaldi<br /> |2009-* (continous updates)<br /> |{{yes|Yes}}<br /> |{{Yes|Apache 2.0}}<br /> |{{some|Windows (not mantained as of 2014), Linux, OSX}}<br /> |[http://kaldi.sourceforge.net/about.html website]<br /> [http://homepages.inf.ed.ac.uk/aghoshal/pubs/asru11-kaldi.pdf paper]<br /> [http://sourceforge.net/p/kaldi/mailman/kaldi-users/ mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://kaldi.sourceforge.net/install.html SVN]<br /> |<br /> |C++<br /> |{{yes|BLAS, LAPACK, GPU (for DNNs)}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMVN<br /> |GMM (Full Cov.), SGMM, DNN<br /> |HLDA, STC, MLLT, MLLR, CMLLR (w/ reg. trees), Exponential transform<br /> |aligment, N-best, lattice rescoring (uses OpenFST)<br /> |Baum-Welch, MMI (boosted), MC, feature-based<br /> |{{yes|Yes}}<br /> |[http://kaldi.sourceforge.net/data_prep.html AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/chime_challenge/WSJ0public/CHiME2012-WSJ0-Kaldi_0.03.tar.gz CHIME-2]<br /> |Weniger2014-REVERB [http://reverb2014.dereverberation.com/workshop/reverb2014-papers/1569884459.pdf Paper] [http://www.mmk.ei.tum.de/~wen/REVERB_2014/kaldi_baseline.tar.gz Code]<br /> |-<br /> !Spraak<br /> |2012 (1.1)<br /> |{{no|No}}<br /> |{{no|[http://www.spraak.org/obtaining-spraak/license proprietary]}}<br /> |{{some|Windows ([http://www.spraak.org/obtaining-spraak/system-requirements limited]), Linux}}<br /> |[http://www.spraak.org/ website]<br /> [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/wambacq/interspeech08/is2008_spraak_v3.pdf paper]<br /> [http://www.spraak.org/mailing-lists mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://www.spraak.org/documentation/doxygen/doc/html/spr__svn.html SVN]<br /> |<br /> |C, Python<br /> |{{no|No}} <br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN,CMN, [http://www.spraak.org/documentation/doxygen/doc/html/spr__tut__mida.html MIDA], [http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__intro.html MDT Techniques]<br /> |GMM, Tied-Mix, Exemplar based<br /> |CMLLR<br /> |aligment, N-best, lattice rescoring, parallel latices<br /> |Baum-Welch<br /> |{{dunno}}<br /> |[http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__example.html AURORA4]<br /> |<br /> |}<br /> <br /> == [[Speaker identification and verification]] ==<br /> <br /> == [[Speech enhancement and separation]] ==<br /> <br /> == [[Other applications]] ==<br /> <br /> == Contribute software ==<br /> To contribute new software, please<br /> * [[Special:UserLogin|create an account]] and login<br /> * go to the wiki page above corresponding to your application; if it does not exist yet, you may create it<br /> * click on the &quot;Edit&quot; link at the top of the page and add a new section for your software (software is ordered by year of the latest version)<br /> * click on the &quot;Save page&quot; link at the bottom of the page to save your modifications<br /> <br /> Please make sure to provide the following information:<br /> * name of the software and year of the latest version<br /> * authors, institution, contact information<br /> * link to the software, ideally including a short demo, and to the external libraries needed<br /> * short description (functionalities, inputs and outputs, programming language, operating system, license, etc) and link to a paper/report describing the software, if any<br /> * whether running on well-known baselines (Aurora-2, Aurora-4, Switchboard, CHiME, etc) is included or requires wrapping by the user<br /> <br /> In order to save storage space, please do not upload the software on this wiki, but link it as much as possible from a public repository (e.g., bitbucket, github, sourceforge) or from a stable URL on the website of your institution. If this is not possible, please contact the [[Main_Page#Working group contacts|resources sharing working group]].</div> Ramon.astudillo https://wiki.inria.fr/wikis/rosp/index.php?title=Software&diff=264 Software 2014-09-08T13:12:42Z <p>Ramon.astudillo: /* Automatic speech recognition */</p> <hr /> <div>This page provides software grouped by application. <br /> <br /> == [[Automatic speech recognition]] ==<br /> <br /> {| class=&quot;wikitable sortable&quot; style=&quot;font-size:72%; border:gray solid 1px; text-align:center; width:auto; table-layout:fixed;&quot;<br /> |-<br /> !style=&quot;width: 40px&quot; rowspan=&quot;2&quot; class=&quot;unsortable&quot;| ASR engines<br /> !colspan=&quot;6&quot; |General attributes<br /> !colspan=&quot;2&quot; |Programming<br /> !colspan=&quot;8&quot; |Implemented ASR techniques<br /> !colspan=&quot;2&quot; |Reproducible research<br /> |-<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | release / update<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | actively developed<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | licence<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | platforms<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | links <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | extensions<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | language<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | hardware optimization<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | VAD<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic features<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | feature normalization / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | acoustic models<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | model adaptation / compensation<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | decoding techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | training techniques<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | online ASR<br /> !scope=&quot;col&quot; width=&quot;40px&quot; | robust ASR training recipes <br /> !scope=&quot;col&quot; width=&quot;40px&quot; | reproducible results<br /> |-<br /> !CMU Sphinx<br /> |1986-* (Sphinx 4.1.0, pocketsphinx 0.8) <br /> |{{yes|Yes}}<br /> |{{some|[https://raw.githubusercontent.com/cmusphinx/sphinx4/master/license.terms proprietary], allows modif.}}<br /> |{{yes|Windows, Linux, OSX (Sphinx4) / Raspberry-pi, iPhone, Android (pocketsphinx)}}<br /> |[http://cmusphinx.sourceforge.net/ website] <br /> [http://www.cs.cmu.edu/~rsingh/homepage/papers/icassp03-sphinx4_2.pdf paper]<br /> [https://www.cs.cmu.edu/~awb/papers/ICASSP2006/0100185.pdf paper]<br /> [http://sourceforge.net/p/cmusphinx/mailman/ mail-list]<br /> [http://sourceforge.net/p/cmusphinx/discussion/ forum]<br /> [https://github.com/cmusphinx github]<br /> |<br /> |Java (Sphinx4), C (pocketsphinx)<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |CMN, Mel-Spectrum subtraction<br /> |GMM, Streams<br /> |MLLR, MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch<br /> |{{yes|Yes}}<br /> |[http://www.keithv.com/software/sphinx/ AURORA4 (WSJ0)]<br /> |<br /> |-<br /> !HTK<br /> |1993-2009 (3.4.1) <br /> |{{no|No}}<br /> |{{no|[http://htk.eng.cam.ac.uk/docs/license.shtml proprietary]}}<br /> |{{yes|Windows, Linux, OSX}}<br /> |[http://htk.eng.cam.ac.uk/download.shtml website]<br /> [http://htk.eng.cam.ac.uk/docs/docs.shtml book]<br /> [http://htk.eng.cam.ac.uk/mailing/subscribe_mail.shtml mail-list]<br /> |[http://htk.eng.cam.ac.uk/extensions/index.shtml official], [http://htk.eng.cam.ac.uk/develop/atk.shtml ATK], [https://github.com/ramon-astudillo/custom_fe uncertainty decoding]<br /> |C<br /> |{{no|No}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMN<br /> |GMM (Full Cov.), Tied-Mix, Streams<br /> |HLDA, MLLR (w/ reg. trees), CMLR (w/ adaptive training), MAP<br /> |aligment, N-best, lattice rescoring<br /> |Baum-Welch, MMI, MBR<br /> |{{yes|Yes}}<br /> |[http://catalog.elra.info/product_info.php?cPath=37_40&amp;products_id=693 AURORA2 (purch.)] [http://catalog.elra.info/index.php?cPath=37_40 AURORA3 (purch.)], [http://www.keithv.com/software/htk/ AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/projects/chime/PCC/data/pcchome.tar.gz CHIME-1], [ftp://ftp.dcs.shef.ac.uk/share/spandh/chime_challenge/grid/eval_tools_grid.tgz CHIME-2-I], [http://spandh.dcs.shef.ac.uk/chime_challenge/chime2_task2.html CHIME-2-II],[http://reverb2014.dereverberation.com/tools/REVERB_TOOLS_FOR_ASR_ver2.0.tgz REVERB]<br /> |ETSI-AFE-AURORA2 [http://aurora.hsnr.de/download/Aurora2_afe_v1_1.pdf paper] (see AURORA2 purch.)<br /> |-<br /> !Kaldi<br /> |2009-* (continous updates)<br /> |{{yes|Yes}}<br /> |{{Yes|Apache 2.0}}<br /> |{{some|Windows (not mantained as of 2014), Linux, OSX}}<br /> |[http://kaldi.sourceforge.net/about.html website]<br /> [http://homepages.inf.ed.ac.uk/aghoshal/pubs/asru11-kaldi.pdf paper]<br /> [http://sourceforge.net/p/kaldi/mailman/kaldi-users/ mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://kaldi.sourceforge.net/install.html SVN]<br /> |<br /> |C++<br /> |{{yes|BLAS, LAPACK, GPU (for DNNs)}}<br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN, CMVN<br /> |GMM (Full Cov.), SGMM, DNN<br /> |HLDA, STC, MLLT, MLLR, CMLLR (w/ reg. trees), Exponential transform<br /> |aligment, N-best, lattice rescoring (uses OpenFST)<br /> |Baum-Welch, MMI (boosted), MC, feature-based, sequence training<br /> |{{yes|Yes}}<br /> |[http://kaldi.sourceforge.net/data_prep.html AURORA4 (WSJ0)], [http://spandh.dcs.shef.ac.uk/chime_challenge/WSJ0public/CHiME2012-WSJ0-Kaldi_0.03.tar.gz CHIME-2]<br /> |Weniger2014-REVERB [http://reverb2014.dereverberation.com/workshop/reverb2014-papers/1569884459.pdf Paper] [http://www.mmk.ei.tum.de/~wen/REVERB_2014/kaldi_baseline.tar.gz Code]<br /> |-<br /> !Spraak<br /> |2012 (1.1)<br /> |{{no|No}}<br /> |{{no|[http://www.spraak.org/obtaining-spraak/license proprietary]}}<br /> |{{some|Windows ([http://www.spraak.org/obtaining-spraak/system-requirements limited]), Linux}}<br /> |[http://www.spraak.org/ website]<br /> [http://www.esat.kuleuven.be/psi/spraak/cgi-bin/get_file.cgi?/wambacq/interspeech08/is2008_spraak_v3.pdf paper]<br /> [http://www.spraak.org/mailing-lists mail-list]<br /> [http://sourceforge.net/p/kaldi/discussion/ forum]<br /> [http://www.spraak.org/documentation/doxygen/doc/html/spr__svn.html SVN]<br /> |<br /> |C, Python<br /> |{{no|No}} <br /> |{{yes|Yes}}<br /> |MFCC, PLP<br /> |VTLN,CMN, [http://www.spraak.org/documentation/doxygen/doc/html/spr__tut__mida.html MIDA], [http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__intro.html MDT Techniques]<br /> |GMM, Tied-Mix, Exemplar based<br /> |CMLLR<br /> |aligment, N-best, lattice rescoring, parallel latices<br /> |Baum-Welch<br /> |{{dunno}}<br /> |[http://www.spraak.org/documentation/doxygen/doc/html/spr__mdt__example.html AURORA4]<br /> |<br /> |}<br /> <br /> == [[Speaker identification and verification]] ==<br /> <br /> == [[Speech enhancement and separation]] ==<br /> <br /> == [[Other applications]] ==<br /> <br /> == Contribute software ==<br /> To contribute new software, please<br /> * [[Special:UserLogin|create an account]] and login<br /> * go to the wiki page above corresponding to your application; if it does not exist yet, you may create it<br /> * click on the &quot;Edit&quot; link at the top of the page and add a new section for your software (software is ordered by year of the latest version)<br /> * click on the &quot;Save page&quot; link at the bottom of the page to save your modifications<br /> <br /> Please make sure to provide the following information:<br /> * name of the software and year of the latest version<br /> * authors, institution, contact information<br /> * link to the software, ideally including a short demo, and to the external libraries needed<br /> * short description (functionalities, inputs and outputs, programming language, operating system, license, etc) and link to a paper/report describing the software, if any<br /> * whether running on well-known baselines (Aurora-2, Aurora-4, Switchboard, CHiME, etc) is included or requires wrapping by the user<br /> <br /> In order to save storage space, please do not upload the software on this wiki, but link it as much as possible from a public repository (e.g., bitbucket, github, sourceforge) or from a stable URL on the website of your institution. If this is not possible, please contact the [[Main_Page#Working group contacts|resources sharing working group]].</div> Ramon.astudillo