|
About the book | Software | Chapter Resources | Corpora | Links
About the book
Human listeners are able to perceptually segregate one sound source from an acoustic mixture, such as a single voice from a mixture of other voices and music at a busy cocktail party. How can we engineer "machine listening" systems that achieve this perceptual feat?
Albert Bregman's book Auditory Scene Analysis, published in 1990, draws an analogy between the perception of auditory scenes and visual scenes, and describes a coherent framework for understanding the perceptual organization of sound. His account has stimulated much interest in computational studies of hearing. Such studies are motivated in part by the demand for practical sound separation systems, which have many applications including noise-robust automatic speech recognition, hearing prostheses, and automatic music transcription. This emerging field has become known as computational auditory scene analysis (CASA).
Computational Auditory Scene Analysis: Principles, Algorithms and Applications provides a comprehensive and coherent account of the state-of-the art in CASA, in terms of the underlying principles, the algorithms and system architectures that are employed, and the potential applications of this exciting new technology. The text is written at a level that will be accessible to graduate students and researchers from related science and engineering disciplines. The extensive bibliography accompanying each chapter will also make this book a valuable reference source.
The book is edited by DeLiang Wang (The Ohio State University, USA) and Guy J. Brown (University of Sheffield, UK), with the following contributing authors:
- Jon Barker (University of Sheffield, UK)
- Alain de Cheveigné (Université Paris 5, France)
- Daniel P. W. Ellis (Columbia University, USA)
- Albert S. Feng (University of Illinois at Urbana-Champaign, USA)
- Masataka Goto (National Institute of Advanced Industrial Science and Technology (AIST), Japan)
- Douglas L. Jones (University of Illinois at Urbana-Champaign, USA)
- Kalle Palomäki (Helsinki University of Technology, Finland)
- Richard Stern (Carnegie Mellon University, USA)
For more information, and to purchase, see amazon.com or amazon.co.uk
Software
Software for producing many of the figures in the book is available for download below:
code.zip [156 KB]
This will unzip into a directory structure that contains a common library (which you should add to your Matlab path) and several directories containing Matlab m-files for specific chapters.
Chapter Resources
Chapter 1. Fundamentals of computational auditory scene analysis
by DeLiang Wang and Guy J. Brown
Software
Matlab m-files to produce the following figures are included in the chapter1 directory of the software archive:
- chap1fig6.m - both panels of figure 1.6
- chap1fig7a.m - left panel of figure 1.7
- chap1fig7b.m - right panel of figure 1.7
- chap1fig9a.m - left panel of figure 1.9
- chap1fig9b.m - right panel of figure 1.9
Matlab software for generating a cochleagram and resynthesizing an audio waveform from the
cochleagram are available here (under "Cochleagram Analysis and Resynthesis").
Chapter 2. Multiple F0 estimation
by Alain de Cheveigné
Supplementary material will be available shortly.
Chapter 3. Feature-based speech segregation
by DeLiang Wang
Software
- Feature extraction. Matlab software for onset/offset detection and envelope extraction is available for download below:
onsetOffset.zip [7.5KB]
envelopeExtraction.zip [6.1KB]
- Auditory segmentation. Software for cross-channel correlation based auditory segmentation is available here
(under "Speech Segregation Based on Oscillatory Correlation").
Software for onset/offset based segmentation is available here (under "Auditory Segmentation").
- Voiced speech segregation. Software for the Hu-Wang model of voiced speech segregation is available here (under "Voiced Speech Segregation").
Audio demonstrations
An audio demo of the Hu-Wang system for monaural voiced speech segregation is available here.
Chapter 4. Model-based scene analysis
by Daniel P. W. Ellis
Supplementary material will be available shortly.
Chapter 5. Binaural sound localization
by Richard M. Stern, Guy J. Brown and DeLiang Wang
Papers and software downloads are available from Richard Stern's web pages for binaural hearing.
Chapter 6. Localization-based grouping
by Albert S. Feng and Douglas L. Jones
Audio demonstrations
Demonstration of the two-sensor FMV system of Lockwood et al. The test condition comprised a target source at 0 degrees azimuth and four interferers at -20, +20, -60 and +60 degrees azimuth. The combined noise had an overall SNR of -4 dB at the position of the two-sensor array, at 1m away from all the sound sources.
Chapter 7. Reverberation
by Guy J. Brown and Kalle J. Palomäki
Audio demonstrations
Harmonic dereverberation (HERB). The following two sets of sound files show the performance of the HERB algorithm for a male and female voice. The female sound files were used to generate figure 7.9. Many thanks to Tomohiro Nakatani for supplying these sound files and allowing us to distribute them.
Chapter 8. Analysis of musical audio signals
by Masataka Goto
Supplementary material for this chapter is available here.
Chapter 9. Robust automatic speech recognition
by Jon Barker
Software
Matlab m-files to produce the following figures are included in chapter9 of the software archive:
- demo1.m - makes a colour version of figure 9.2
- demo2.m - an interactive version of figure 9.3
This directory also includes figure 9.1 in electronic form (fig1.eps).
Jon Barker's CASA toolkit (CTK) is available here.
Chapter 10. Neural and perceptual modeling
by Guy J. Brown and DeLiang Wang
Software
See the MAD demonstrations wangNeuron and wangNetwork for simulations of a single neural oscillator and a small oscillator network, respectively.
Matlab m-files to produce the following figures are included in the chapter10 directory of the software archive:
- chap10figs7and8.m - all panels of figure 10.7, and panels B and C of figure 10.8
The complete C source code for the Wang-Brown model of speech segregation based on oscillatory correlation
is available here (under "Speech Segregation Based on Oscillatory Correlation").
Corpora
Links
Software tools
- DSAM - Development system for auditory modelling (University of Essex)
- Malcolm Slaney's auditory toolbox
- MAD - Matlab auditory demonstrations (University of Sheffield)
- AIM - Auditory Image Model (University of Cambridge)
- Roomsim - Matlab application for sound spatialisation in a rectangular room
- CTK - Jon Barker's CASA toolkit
Web sites about auditory perception and CASA
- www.auditory.org - home page for the AUDITORY list, an email list for the discussion of organizational aspects of auditory perception
Research groups
Personal web pages
If you are working in the field of CASA and would like your web page added to this list, please email us.
- Barker, Jon (University of Sheffield)
- Brown, Guy J. (University of Sheffield)
- de Cheveigné, Alain (Université Paris 5)
- Cooke, Martin (University of Sheffield)
- Ellis, Dan (Columbia University)
- Feng, Albert (Beckman Institute/University of Illinois)
- Goto, Masataka (AIST, Japan)
- Irino, Toshio (Wakayama University)
- Jones, Douglas L. (University of Illinois at Urbana-Champaign)
- Kawahara, Hideki (Wakayama University)
- Lazzaro, John (UC Berkeley)
- Lyon, Richard F. (Dick)
- Okuno, Hiroshi (Kyoto University)
- Palomäki, Kalle (Helsinki University of Technology)
- Patterson, Roy (University of Cambridge)
- Pichevar, Ramin (Communications Research Centre, Ottawa/University of Sherbrooke)
- Rouat, Jean (Université de Montréal, Université de Sherbrooke)
- Slaney, Malcolm (Yahoo! Research)
- Stern, Richard (Carnegie Mellon University)
- Wang, DeLiang (Ohio State University)
- Wrigley, Stuart (University of Sheffield)
Last updated by Guy on 28/6/2007
The keywords for this site are computational auditory scene analysis, CASA, auditory perception, machine hearing, Guy Brown, DeLiang Wang, Wiley, IEEE press, book, textbook, auditory model, sound separation, independent component analysis, ICA, matlab, source code, sound example, audio demonstration, hearing aid, Albert Bregman
|
