ICML 2009 Workshop on Sparse Methods for Music Audio

The workshop is in response to recent developments in the use of sparse coding for audio analysis, but this will be interpreted in the broadest possible light in organizing the workshop.

Workshop organized in conjunction with the 26th International Conference on Machine Learning
Montreal, June 14-18, 2009

DESCRIPTION
Sparse coding is gaining attention as alternative to traditional, orthogonal-basis  approaches, able to find more interesting or more useful solutions to underconstrained, high-dimensional problems. Music audio provides an excellent candidate for sparse coding, being very high dimensional (e.g. over 80,000 values in one second of music from a CD), yet usefully described as the combination of a small number of separate signals -- such as individual instruments -- each subject to a large number of mutual constraints. This description can be applied at multiple levels, from the raw audio through to compositional structure.

There have been only a few publications on applying sparse techniques in music. The goal of the workshop is to bring together researchers with an interest in this topic, to focus, develop, and refine the various perspectives and approaches possible. We hope to raise the profile of these ideas, both to those already working with music audio, and to machine learning researchers who may be curious about working with music audio data.

TOPICS
Since this is an exploratory workshop, we solicit contributions that relate to sparse methods in music audio, interpreted in the broadest possible sense.

Particular topics include audio classification, music recommendation
music structure discovery/recovery, polyphonic pitch extraction, scene analysis/source separation, measures of music similarity,
auditory perception models, music audio representation and coding.

ORGANIZERS

Douglas Eck 
Associate Professor, University of Montreal Computer Science
LISA Machine Learning Lab / BRAMS International Laboratory for Brain, Music and Sound Research

Dan Ellis 
Associate Professor, Electrical Engineering, Columbia University

LabROSA Laboratory for the Recognition and Organization of Speech and Audio

 

Philippe Hamel

PhD Candidate, University of Montreal Computer Science
LISA Machine Learning Lab / BRAMS International Laboratory for Brain, Music and Sound Research

This event will be held in English