[AUDITORY] Final registration reminder: CHiME 2013, 2nd International Workshop on Machine Listening in Multisource Environments (jon )

Subject: [AUDITORY] Final registration reminder: CHiME 2013, 2nd International Workshop on Machine Listening in Multisource Environments From: jon <j.barker@xxxxxxxx> Date: Tue, 30 Apr 2013 12:07:38 +0100 List-Archive:<http://lists.mcgill.ca/scripts/wa.exe?LIST=AUDITORY> --Apple-Mail=_411A069F-771C-4335-AD39-B9344C38D804 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=iso-8859-1 ----------------------------------------------=20 2nd International Workshop on=20 Machine Listening in Multisource Environments (CHiME 2013)=20 in conjunction with ICASSP 2013=20 June 1, 2013, Vancouver, Canada=20 http://spandh.dcs.shef.ac.uk/chime_workshop/=20 ----------------------------------------------=20 *REGISTRATION DEADLINE*: May 11, 2013 (only 2 weeks left!)=20 *FINAL PROGRAMME*=20 http://spandh.dcs.shef.ac.uk/chime_workshop/programme.html=20 *KEYNOTES*=20 Model-based Speech Separation and Recognition: Yesterday, Today, and=20 Tomorrow=20 Steven J. Rennie, IBM=20 Recently, model-based approaches for multi-talker speech separation and=20= recognition have demonstrated great success in highly constrained=20 scenarios, and efficient algorithms for separating data with literally=20= *trillions* of underlying states have been unveiled. In less constrained=20= scenarios, deep neural networks (DNNs) learned on features inspired by=20= human auditory processing have shown great capacity for directly=20 learning masking functions from parallel data. Ideally, a robust speech=20= separation/recognition system should be continuously learning, adapting,=20= and exploiting structure that is present in both target and peripheral=20= signals and interactions, make minimal assumptions about the data to be=20= separated/recognized, not require parallel data streams, and have=20 essentially unlimited information capacity. In this talk I=EF=BF=BDll = briefly=20 review the current state of robust speech separation/recognition=20 technology--where we are, where we apparently need to go, and how we=20 might get there. I'll then discuss in more detail recent work that I've=20= been involved with that is aligned with these goals. Specifically, I=20 will discuss some new results on efficiently learning the structure of=20= models and efficiently optimizing a wide class of matrix-valued=20 functions, some recent work on Factorial Restricted Boltzmann machines=20= for robust ASR, and finally, Direct product DBNs, a new architecture=20 that makes it feasible to learn DNNs with literally *millions* of = neurons.=20 Recognizing and Classifying Environmental Sounds=20 Daniel P.W. Ellis, Columbia University=20 Animal hearing exists to extract useful information out of the=20 environment, and for a lot of animals for a large portion of the=20 evolutionary history of hearing this sound environment has not consisted=20= of speech or music, but of more generic acoustic information arising=20 from collisions, motions, and other events in the external world. This=20= aspect of sound analysis -- getting information out of non-speech,=20 non-music, environmental sounds -- is finally beginning to gain=20 popularity in research since it holds promise as a tool for automatic=20 search and retrieval of audio/video recordings, an increasingly urgent=20= problem. I will discuss our recent work in using audio analysis to=20 manage and search environmental sound archives (including personal audio=20= lifelogs and consumer video collections), and illustrate with some of=20 the approaches that work more or less well, with an effort to explain = why.=20 *OVERVIEW*=20 CHiME 2013 will consider the challenge of developing machine listening=20= applications for operation in multisource environments, i.e. real-world=20= conditions with acoustic clutter, where the number and nature of the=20 sound sources is unknown and changing over time. It will bring together=20= researchers from a broad range of disciplines (computational hearing,=20 blind source separation, speech recognition, machine learning) to=20 discuss novel and established approaches to this problem. The=20 cross-fertilisation of ideas will foster fresh approaches that=20 efficiently combine the complementary strengths of each research field.=20= One highlight of the Workshop will be the presentation of the results of=20= the 2nd CHiME Speech Separation and Recognition Challenge, that is a=20 two-microphone multisource speech separation and recognition challenge=20= supported by the IEEE AASP, MLSP and SLTC Technical Committees. To find=20= out more, please visit http://spandh.dcs.shef.ac.uk/chime_challenge.=20 *REGISTRATION*=20 To register, please visit=20 http://spandh.dcs.shef.ac.uk/chime_workshop/registration.html=20 The registration fee is 35 UK pounds and includes admission to the=20 sessions, electronic proceedings, buffet lunch, and tee and coffee = breaks.=20 *VENUE*=20 The workshop is taking place at the Hyatt Regency Vancouver, 655 Burrard=20= Street -- close to the ICASSP 2013 venue -- on the day after ICASSP=20 finishes, Saturday 1st June. Information about accommodation and how to=20= get to and from downtown Vancouver can be found on the main ICASSP = website:=20 http://www.icassp2013.com=20 See you in Vancouver.=20 Best regards,=20 CHiME Organising Committee=20 --Apple-Mail=_411A069F-771C-4335-AD39-B9344C38D804 Content-Transfer-Encoding: quoted-printable Content-Type: text/html; charset=iso-8859-1 <html><head><meta http-equiv=3D"Content-Type" content=3D"text/html = charset=3Diso-8859-1"></head><body style=3D"word-wrap: break-word; = -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; ">      ----------------------------------------------          =        2nd International Workshop on  Machine Listening in Multisource Environments (CHiME = 2013)          =       in conjunction with ICASSP 2013                June 1, 2013, = Vancouver, Canada        =   <a = href=3D"http://spandh.dcs.shef.ac.uk/chime_workshop/" target=3D"_blank" = style=3D"margin: 0px; padding: 0px; border: 0px; vertical-align: = baseline; text-decoration: none; color: rgb(102, 17, 204); cursor: = pointer; font-family: Arial, Helvetica, sans-serif; font-size: 13px; = background-color: rgb(255, 255, 255); = ">http://spandh.dcs.shef.ac.uk/chime_workshop/</a>        =  ----------------------------------------------  *REGISTRATION DEADLINE*: May 11, = 2013 (only 2 weeks left!)  *FINAL = PROGRAMME*  <a = href=3D"http://spandh.dcs.shef.ac.uk/chime_workshop/programme.html" = target=3D"_blank" style=3D"margin: 0px; padding: 0px; border: 0px; = vertical-align: baseline; text-decoration: none; color: rgb(102, 17, = 204); cursor: pointer; font-family: Arial, Helvetica, sans-serif; = font-size: 13px; background-color: rgb(255, 255, 255); = ">http://spandh.dcs.shef.ac.uk/chime_workshop/programme.html</a><span= style=3D"color: rgb(34, 34, 34); font-family: Arial, Helvetica, = sans-serif; font-size: 13px; background-color: rgb(255, 255, 255); = ">  *KEYNOTES*  Model-based Speech Separation = and Recognition: Yesterday, Today, and  Tomorrow  Steven J. Rennie, IBM  Recently, model-based approaches = for multi-talker speech separation and  recognition have = demonstrated great success in highly constrained  scenarios, and efficient algorithms for separating data with = literally  *trillions* of underlying states = have been unveiled. In less constrained  scenarios, deep neural = networks (DNNs) learned on features inspired by  human auditory processing have shown great capacity for = directly  learning masking functions from = parallel data. Ideally, a robust speech  separation/recognition = system should be continuously learning, adapting,  and = exploiting structure that is present in both target and = peripheral  signals and interactions, make = minimal assumptions about the data to be  separated/recognized, not = require parallel data streams, and have  essentially unlimited = information capacity. In this talk I=EF=BF=BDll briefly  review the current state of robust speech = separation/recognition  technology--where we are, where = we apparently need to go, and how we  might get there. I'll then = discuss in more detail recent work that I've  been involved with that is aligned with these goals. Specifically, = I  will discuss some new results on efficiently learning the = structure of  models and efficiently = optimizing a wide class of matrix-valued  functions, some recent = work on Factorial Restricted Boltzmann machines  for = robust ASR, and finally, Direct product DBNs, a new = architecture  that makes it feasible to learn = DNNs with literally *millions* of neurons.  Recognizing and = Classifying Environmental Sounds  Daniel P.W. Ellis, Columbia = University  Animal hearing exists to extract = useful information out of the  environment, and for a lot of = animals for a large portion of the  evolutionary history of = hearing this sound environment has not consisted  of = speech or music, but of more generic acoustic information = arising  from collisions, motions, and = other events in the external world.  This  aspect of sound analysis -- getting information out of = non-speech,  non-music, environmental sounds = -- is finally beginning to gain  popularity in research since it = holds promise as a tool for automatic  search and retrieval of = audio/video recordings, an increasingly urgent  problem.  I will discuss our recent work in using audio analysis = to  manage and search environmental sound archives (including = personal audio  lifelogs and consumer video = collections), and illustrate with some of  the approaches that work = more or less well, with an effort to explain why.  *OVERVIEW*  CHiME 2013 will consider the = challenge of developing machine listening  applications for operation = in multisource environments, i.e. real-world  conditions with acoustic clutter, where the number and nature of = the  sound sources is unknown and changing over time. It will bring = together  researchers from a broad range = of disciplines (computational hearing,  blind source separation, = speech recognition, machine learning) to  discuss novel and = established approaches to this problem. The  cross-fertilisation of ideas will foster fresh approaches = that  efficiently combine the = complementary strengths of each research field.  One = highlight of the Workshop will be the presentation of the results = of  the 2nd CHiME Speech Separation and Recognition Challenge, that = is a  two-microphone multisource = speech separation and recognition challenge  supported by the IEEE AASP, MLSP and SLTC Technical Committees. To = find  out more, please = visit <a href=3D"http://spandh.dcs.shef.ac.uk/chime_challenge"= target=3D"_blank" style=3D"margin: 0px; padding: 0px; border: 0px; = vertical-align: baseline; text-decoration: none; color: rgb(102, 17, = 204); cursor: pointer; font-family: Arial, Helvetica, sans-serif; = font-size: 13px; background-color: rgb(255, 255, 255); = ">http://spandh.dcs.shef.ac.uk/chime_challenge</a>.  *REGISTRATION*  To register, please = visit  <a = href=3D"http://spandh.dcs.shef.ac.uk/chime_workshop/registration.html" = target=3D"_blank" style=3D"margin: 0px; padding: 0px; border: 0px; = vertical-align: baseline; text-decoration: none; color: rgb(102, 17, = 204); cursor: pointer; font-family: Arial, Helvetica, sans-serif; = font-size: 13px; background-color: rgb(255, 255, 255); = ">http://spandh.dcs.shef.ac.uk/chime_workshop/registration.html<= /a>  The registration fee is 35 UK pounds and includes admission to = the  sessions, electronic proceedings, buffet lunch, and tee and = coffee breaks.  *VENUE*  The = workshop is taking place at the Hyatt Regency Vancouver, 655 = Burrard  Street -- close to the ICASSP = 2013 venue -- on the day after ICASSP  finishes, Saturday 1st = June. Information about accommodation and how to  get = to and from downtown Vancouver can be found on the main ICASSP = website:  <a href=3D"http://www.icassp2013.com/" = target=3D"_blank" style=3D"margin: 0px; padding: 0px; border: 0px; = vertical-align: baseline; text-decoration: none; color: rgb(102, 17, = 204); cursor: pointer; font-family: Arial, Helvetica, sans-serif; = font-size: 13px; background-color: rgb(255, 255, 255); = ">http://www.icassp2013.com</a>  See you in = Vancouver.  Best regards,  CHiME Organising Committee  <div = apple-content-edited=3D"true"> <span class=3D"Apple-style-span" = style=3D"border-collapse: separate; color: rgb(0, 0, 0); font-family: = Helvetica; font-size: medium; font-style: normal; font-variant: normal; = font-weight: normal; letter-spacing: normal; line-height: normal; = orphans: 2; text-indent: 0px; text-transform: none; white-space: normal; = widows: 2; word-spacing: 0px; -webkit-border-horizontal-spacing: 0px; = -webkit-border-vertical-spacing: 0px; = -webkit-text-decorations-in-effect: none; -webkit-text-size-adjust: = auto; -webkit-text-stroke-width: 0px; "><div style=3D"word-wrap: = break-word; -webkit-nbsp-mode: space; -webkit-line-break: = after-white-space; "> </div></div></body></html>= --Apple-Mail=_411A069F-771C-4335-AD39-B9344C38D804--

This message came from the mail archive
/var/www/postings/2013/
maintained by:

DAn Ellis <dpwe@ee.columbia.edu>
Electrical Engineering Dept., Columbia University