Sparse approximation attempts to model a signal using few terms selected from a large set of functions (atoms) that can be defined with minimal restrictions. Even though research in this area began more than 15 years ago within the signal processing field, the computational costs associated with applying sparse approximation to audio data have remained prohibitive until recently. Thus, such applications are just now beginning to be explored. In my work, I specifically address the use of sparse approximation to facilitate representations of audio data to provide a rich interface to the contents of such signals for purposes of analysis, extraction, and modification. This talk will present an overview of such work.

Bob has received an undergraduate degree in physics from the University of Colorado, Boulder (B.A. 1998), a graduate degree in computer music from Stanford University (M.A. 1999), and a few other graduate degrees from the University of California, Santa Barbara (M.S. 2004, M.S. 2007, Ph. D. 2009). He continues his research in sparse approximation and signal representation as a Chateaubriand Fellow post-doctoral researcher at UPMC - Paris 06 with Professor Laurent Daudet.

Georg Essl, georg.essl(AT)telekom.de