Sound Classification by Prof. Dan Ellis (Columbia)
I’m happy to welcome Prof. Dan Ellis (from Columbia, and on sabbatical at Google) to Stanford CCRMA to talk about recognizing environmental sounds. Recognizing speech and music are relatively common applications of machine learning. But what about the rest of the world? Speech and music are only a small fraction of the sounds that we hear throughout our day.
Dan Ellis has been at the intersection of speech recognition, audio analysis and music processing research throughout his career. He brings an unusual range of interests and skills to all three problems, and I highly recommend his work.
Who: Prof. Dan Ellis (Columbia and on Sabbatical at Google Research)
What: Recognizing Environmental Sounds
When: Friday March 13th at 11AM
Where: CCRMA Seminar Room
Why: Because even CCRMA doesn’t live on music alone
Bring your favorite sound recognizer to CCRMA. I’m sure that the quality of the talk will make it easy to ignore the usual Hearing Seminar lawnmower visit.
— Malcolm
Recognizing Environmental Sounds
In addition to communication, hearing is useful for informing us about general events in the world, since any kind of physical contact or abrasion will radiate sound. For computers, however, the automatic detection and recognition of sounds other than speech and music has received relatively little attention. I will briefly review past work in automatic recognition of these so-called environmental sounds, including several recent evaluations.
The current generation of big-data-fueled machine learning classifiers offers the possibility of huge advances over the current state of the art, but demands large numbers of training examples. I'll discuss various responses to this need, look at some results, and consider likely future outcomes.
Daniel P. W. Ellis received the Ph.D. degree in electrical engineering from theMassachusetts Institute of Technology, Cambridge, where he was a Research Assistant in the Machine Listening Group of the Media Lab. He spent several years as a Research Scientist at the International Computer Science Institute, Berkeley, CA. Currently, he is a Full Professor with the Electrical Engineering Department, Columbia University, New York. His Laboratory for Recognition and Organization of Speech and Audio (LabROSA) is concerned with all aspects of extracting high-level information from audio, including speech recognition, music description, and environmental sound processing. He also runs the AUDITORYemail list of 2100 worldwide researchers in perception and cognition of sound.
Dan Ellis is currently with the Sound Understanding group at Google while on sabbatical from Columbia University.