A team from the Ming Hsieh Department of Electrical Engineering's Signal Analysis and Interpretation Laboratory working with psychology faculty received a 2010 Best Paper Award at the International Speech Communication Association (ISCA) Interspeech 2010 annual conference in Makuhari Japan
A team led by Panayiotis Georgiou and Shrikanth Narayanan analyzed interactions between married couples
It analyzed a corpus of conversations of married couples spontaneously interacting about a problem in their relationship. The goal was to classify the spouses' behavior using features derived from the audio signal. Based on automatic segmentation, the team extracted prosodic/spectral features to capture global acoustic properties for each spouse, and using machine learning techniques, they trained classifiers to predict the behavior of each spouse for six codes.
This effort is part of a larger endeavor involving Behavioral Signal Processing, which resulted in two other related Viterbi School-linked papers at the same conference: "Quantification of Prosodic Entrainment in Affective Spontaneous Spoken Interactions of Married Couples" and "A New Multichannel Multimodal Dyadic Interaction Database."
The long-term goal focuses on mental health research and practice; it involves employing signal processing and machine learning technologies to sense human behaviors that aid in and transform observational methods. Funding for the work came from the National Science Foundation and the Viterbi Research Innovation Fund.
"Our department joins me in congratulating Panos Georgiou, Shri Narayanan and their students for their 2010 Best Paper Award at the Interspeech 2010 conference,” said Alexander (Sandy) Sawchuk, Chair of the Ming Hsieh EE-Systems Department. “It is a significant appreciation of their work in behavioral signal processing by the research community."
More than 1,000 participants from all over the world attend the Interspeech conference each year, and more than 700 papers are presented. Best Paper Awardees are invited to submit expanded versions of their work to the prestigious Speech Communication and Computer Science and Language journals.
EE Ph.D. student Matthew Black was author of the prizewinning paper
Matthew Black, the lead author, is a PhD candidate in EE, whose research interests are in behavioral signal processing, specifically in the automatic quantification and emulation of human observational processes to describe human behavior. He is a recipient of the Alfred E. Mann Innovation in Engineering Doctoral Fellowship 2010-2011 and the Simon Ramo Scholarship 2009-2010 at USC.
Panayiotis Georgiou received his MS and PhD degrees from USC, and is now a research assistant professor. He has been a member of the Speech Analysis and Interpretation Lab since 2003, and has published over 60 papers in the fields of statistical signal processing, alpha stable distributions, speech and multimodal signal processing and interfaces, speech translation, language modeling, immersive sound processing, sound source localization, and speaker identification. He has been an Investigator and co-PI on several federally funded projects, notably the DARPA Transtac "SpeechLinks" and the NSF "An Integrated Approach to Creating Enriched Speech Translation Systems." He is currently serving as guest editor of Computer Speech and Language.
Shrikanth Narayanan is professor of electrical engineering, computer science, linguistics and psychology and the Andrew J. Viterbi Professor of Engineering. He also directs the Signal Analysis and Interpretation Laboratory. Narayanan has received many professional and technical awards and honors for his work in speech and language processing, analysis, recognition, modeling, and synthesis, and for work in biomedical signal processing, human-machine interaction and multimodal interfaces.
The International Speech Communication Association promotes, in an international worldwide context, activities and exchanges in all fields related to speech communication science and technology. The association is aimed at all persons and institutions interested in fundamental research and technological development that aims at describing, explaining and reproducing the various aspects of human communication by speech, that is, without assuming this enumeration to be exhaustive, phonetics, linguistics, computer speech recognition and synthesis, speech compression, speaker recognition, aids to medical diagnosis of voice pathologies.