         
|
GRASP Lab Seminar 2003-2004
September 26, 11:00 AM, Levine Hall 307, hosted by Lawrence Saul.
Dan Ellis
Columbia University
Sound, Mixtures, and Learning: LabROSA Overview
Abstract: Human listeners, like other animals, gain a great deal of information from their acoustic environment. Sound provides a useful and complementary information channel for situations in which vision is inadequate, e.g., for detecting events regardless of their direction, and for operating in darkness. As with visual scenes, however, the presence of multiple, interfering sources makes the extraction of reliable, high-level information from real-world sounds extremely challenging.
The most successful application of acoustic analysis is automatic speech recognition; however, well-performing speech recognizers require highly-controlled acoustic environments and rely on the assumption that the signal is dominated by a single voice. Computational auditory scene analysis (CASA) is concerned with separating different sound sources by copying what we understand of human performance. Other ideas from independent component analysis and machine learning have also been applied to this problem.Biography: Prior to founding LabROSA at Columbia in 2000, Dan Ellis was a senior research scientist at the International Computer Science Institute in Berkeley CA, working on robust speech recognition. His Ph.D. from the MIT Media Lab is in models of human sound organization. Current LabROSA projects include speech recognition and information extraction from meeting room audio, general nonspeech sound mixture analysis, and musical sound analysis, classification, and similarity-based recommendation. Prof. Ellis also manages the AUDITORY email list for researchers interested in perceptual organization of sound, which was founded in 1992.
