Subject: 2nd call: CHiME Speech Separation and Recognition Challenge From: Jon Barker <j.barker@xxxxxxxx> Date: Tue, 30 Oct 2012 12:00:40 +0000 List-Archive:<http://lists.mcgill.ca/scripts/wa.exe?LIST=AUDITORY>--Apple-Mail=_873152C1-8405-4E69-A39B-50429EBB5640 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=us-ascii ---------------------------------------------- 2nd CHiME Speech Separation and Recognition Challenge Supported by the IEEE AASP, MLSP and SLTC Technical Committees Deadline: January 15, 2013 Workshop: June 1, 2013, Vancouver, Canada http://spandh.dcs.shef.ac.uk/chime_challenge/ ---------------------------------------------- NEWS The *test datasets* have been released on the website! Noise-free = datasets have also been made available for benchmarking purposes. Additional *instructions* have been added in order to make the systems = more comparable and facilitate scientific analysis of the results. Check = the website for details. We are proud to announce one additional *sponsor* for our workshop. The = best challenge papers will receive a monetary prize. We are envisaging a *special issue* of IEEE Signal Processing Magazine = to which the authors of the best challenge papers will be invited to = contribute. FEATURED TASKS The challenge consists of recognising distant-microphone speech mixed in two-channel nonstationary noise recorded over a period of several weeks in a real family house. Entrants may address either one or both of the following tracks: Medium vocabulary track: WSJ 5k sentences uttered by a static speaker Small vocabulary track: simpler commands but small head movements TO ENTER You will find everything you need to get started (and even more) on the challenge website: - a full description of the challenge, - clean, reverberated and multi-condition training and development data, - baseline training, decoding and scoring software tools based on HTK. Submission consists of a 2- to 8-page paper describing your system and reporting its performance on the development and test datasets. In addition, you are welcome to submit an earlier paper to ICASSP 2013, which will tentatively be grouped with other papers into a dedicated session. Any approach is welcome, whether emerging or established. If you are interested in participating, please email us so we can monitor interest and send you further updates about the challenge. BEST CHALLENGE PAPER AWARDS The best challenge papers will distinguished by an award from the Industrial Board. IMPORTANT DATES July 2012 Launch October 2012 Test set release January 15, 2013 Challenge & workshop submission deadline February 18, 2013 Paper notification & release of the challenge results June 1, 2013 post-ICASSP workshop INDUSTRIAL BOARD Masami Akamine, Toshiba Carlos Avendano, Audience Li Deng, Microsoft Erik McDermott, Google Gautham Mysore, Adobe Atsushi Nakamura, NTT Peder A. Olsen, IBM Trausti Thormundsson, Conexant Daniel Willett, Nuance WORKSHOP SPONSORS Conexant Systems Inc. Audience Inc. Adobe Systems Inc. Google Inc. Mitsubishi Electric Research Laboratories ORGANISERS Emmanuel Vincent, INRIA Jon Barker, University of Sheffield Shinji Watanabe & Jonathan Le Roux, MERL Francesco Nesta & Marco Matassoni, FBK-IRST --Apple-Mail=_873152C1-8405-4E69-A39B-50429EBB5640 Content-Transfer-Encoding: quoted-printable Content-Type: text/html; charset=us-ascii <html><head><meta http-equiv=3D"Content-Type" content=3D"text/html = charset=3Dus-ascii"></head><body style=3D"word-wrap: break-word; = -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; = "> = ----------------------------------------------<br><br> &n= bsp; 2nd CHiME Speech Separation and Recognition = Challenge<br>Supported by the IEEE AASP, MLSP and SLTC Technical = Committees<br><br> &n= bsp; Deadline: January = 15, = 2013<br> Worksh= op: June 1, 2013, Vancouver, = Canada<br><br> <a = href=3D"http://spandh.dcs.shef.ac.uk/chime_challenge/">http://spandh.dcs.s= hef.ac.uk/chime_challenge/</a><br><br> = ----------------------------------------------<br><br><br>NEWS= <br><br>The *test datasets* have been released on the website! = Noise-free datasets have also been made available for benchmarking = purposes.<br><br>Additional *instructions* have been added in order to = make the systems more comparable and facilitate scientific analysis of = the results. Check the website for details.<br><br>We are proud to = announce one additional *sponsor* for our workshop. The best challenge = papers will receive a monetary prize.<br><br>We are envisaging a = *special issue* of IEEE Signal Processing Magazine to which the authors = of the best challenge papers will be invited to = contribute.<br><br><br>FEATURED TASKS<br><br>The challenge consists of = recognising distant-microphone speech mixed in<br>two-channel = nonstationary noise recorded over a period of several weeks<br>in a real = family house. Entrants may address either one or both of = the<br>following tracks:<br><br>Medium vocabulary track: WSJ 5k = sentences uttered by a static speaker<br><br>Small vocabulary track: = simpler commands but small head movements<br><br><br>TO ENTER<br><br>You = will find everything you need to get started (and even more) on = the<br>challenge website:<br>- a full description of the challenge,<br>- = clean, reverberated and multi-condition training and development = data,<br>- baseline training, decoding and scoring software tools based = on HTK.<br><br>Submission consists of a 2- to 8-page paper describing = your system and<br>reporting its performance on the development and test = datasets. In<br>addition, you are welcome to submit an earlier paper to = ICASSP 2013,<br>which will tentatively be grouped with other papers into = a dedicated<br>session.<br><br>Any approach is welcome, whether emerging = or established.<br><br>If you are interested in participating, please = email us so we can<br>monitor interest and send you further updates = about the challenge.<br><br><br>BEST CHALLENGE PAPER AWARDS<br><br>The = best challenge papers will distinguished by an award from = the<br>Industrial Board.<br><br><br>IMPORTANT DATES<br><br>July 2012 = Launch<br>October = 2012 Test set release<br>January 15, = 2013 Challenge & workshop submission = deadline<br>February 18, 2013 Paper notification & release of = the challenge results<br>June 1, 2013 = post-ICASSP = workshop<br><br><br>INDUSTRIAL BOARD<br><br>Masami Akamine, = Toshiba<br>Carlos Avendano, Audience<br>Li Deng, Microsoft<br>Erik = McDermott, Google<br>Gautham Mysore, Adobe<br>Atsushi Nakamura, = NTT<br>Peder A. Olsen, IBM<br>Trausti Thormundsson, Conexant<br>Daniel = Willett, Nuance<br><br><br>WORKSHOP SPONSORS<br><br>Conexant Systems = Inc.<br>Audience Inc.<br>Adobe Systems Inc.<br>Google Inc.<br>Mitsubishi = Electric Research Laboratories<br><br><br>ORGANISERS<br><br>Emmanuel = Vincent, INRIA<br>Jon Barker, University of Sheffield<br>Shinji Watanabe = & Jonathan Le Roux, MERL<br>Francesco Nesta & Marco Matassoni, = FBK-IRST<br><br></body></html>= --Apple-Mail=_873152C1-8405-4E69-A39B-50429EBB5640--