Colloquium Details
Recognition of unstructured environmental sounds
| Author: | Selina Chu Oregon State University |
|---|---|
| Date: | April 26, 2012 |
| Time: | 15:30 |
| Location: | 220 Deschutes |
| Host: | Daniel Lowd |
Abstract
Recognizing environmental sounds is a basic audio signal processing problem. Human beings utilize both vision and hearing to navigate and respond to their surroundings, a capability still quite limited in machine processing. Consider, for example, applications in assistive robotics, and other mobile device-based services, where context aware processing is often desired. The first step toward achieving this goal is the ability to process unstructured audio and recognize audio scenes (or environments). The aim of this work is on the characterization of unstructured environmental sounds for understanding and predicting the context surrounding of an agent or device. I will talk about some of the challenging issues in characterizing unstructured environmental sounds and the development of appropriate feature extraction algorithm and learning techniques for modeling the variations of the acoustic environment.
In the last part of my talk, I will briefly discuss my current research efforts in a collaboration project between Oregon State University and Cornell Lab of Ornithology to address some issues on sampling bias by incorporating aspects of covariate shift into the species distribution modeling process.
Biography
Selina Chu is a postdoctoral researcher in the School of EECS at Oregon State University and a Computing Innovation Fellow (CI Fellow). She received her PhD from the Computer Science department at University of Southern California in 2011. She was a member of the Speech Analysis and Interpretation Lab and also the Multimedia Communications Lab, where she worked with Prof. Shri Narayanan and Prof. C.C. Jay Kuo. She holds a MS in Information and Computer Science from University of California-Irvine and BS in Electrical Engineering from California State Polytechnic University. She has also worked as a research intern at AT&T Labs-Research and IBM T.J. Watson Research Center. Her previous research has been in the areas of general unstructured audio. Currently she is working on problems in machine learning algorithms for species distribution.