| ------------------------------ 2nd International Workshop on Machine Listening in Multisource Environments (CHiME 2013) in conjunction with ICASSP 2013 June 1, 2013, Vancouver, Canada http://spandh.dcs.shef.ac.uk/ ----------------------------- *REGISTRATION DEADLINE*: May 11, 2013 (only 2 weeks left!) *FINAL PROGRAMME* http://spandh.dcs.shef.ac.uk/ *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�ll 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 http://spandh.dcs.shef.ac.uk/ *REGISTRATION* To register, please visit http://spandh.dcs.shef.ac.uk/ 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: http://www.icassp2013.com See you in Vancouver. Best regards, CHiME Organising Committee |