Faculty Profile: Richard Sproat

The ability to continue the interdisciplinary approach he enjoyed at Bell Labs was one reason Richard Sproat joined the Beckman Institute.

Richard Sproat felt energized by the interdisciplinary research environment he found while working at Bell Labs and AT&T Labs. His desire to continue doing research in that type of setting was one reason Sproat came to the Beckman Institute and the University of Illinois in 2003.

“One of the things that was always very nice about Bell Labs—which was quite different from what I knew the case to be in academia—was that I could walk down the hall and talk to an engineer, or a computer scientist or psychologist, all on the same hallway,” Sproat said. “You would never get that in a typical academic environment. So one of the things that was so appealing about Beckman is you do have people with all kinds of backgrounds and we really do work together. So it was really the interdisciplinary nature that was the pull here.”

What the business world did not quite prepare him for was the juggling act that often comes with being a full-time professor and researcher at a place like Beckman. “One of the things I have to admit is that when I came to academia I didn't really know what the term busy meant,” Sproat said with a laugh. “It's not so much the amount of work; the issue is the number of things that I have to keep juggling at the same time.”

A member of Beckman's Artificial Intelligence group who holds faculty appointments in both the Department of Electrical and Computer Engineering and the Department of Linguistics, Sproat has more than a half-dozen current research projects.

Sproat's research interests center on various aspects of speech processing, as well as writing systems, and all of his work has a computational component to it. His interests include such topics as language modeling for colloquial Arabic speech recognition, named entity detection and transliteration for multiple languages, and prediction of prosody from text for affective speech synthesis.

Sproat said the named entity detection project is one of two he is currently spending a lot of his time on. It involves creating software that can pick out the same words or phrases in streams of news texts from varying languages, including ones with completely different scripts such as Chinese and English.

The other key project, funded by a Critical Research Initiative grant from the University of Illinois, is for developing methods for second-language fluency assessment. That project is part of a collaboration with Beckman colleagues Chilin Shih, Kay Bock, Brian Ross, and Mark Hasegawa-Johnson. They are looking at developing ways to assess and improve second- language fluency, an important topic in our increasingly global economy. There is hope the project will lead to a new center on campus for second language learning.

If it happens, the center would truly be interdisciplinary, Sproat said. Bock is a psycholinguistic, Ross is a psychologist, while Hasegawa-Johnson tackles speech issues from an engineering perspective. Sproat said the second language project and a future center would have both research and practical implications that could lead to a quantitative measure of fluency and improve the teaching of second languages.

“There's a whole bunch of issues out there to be investigated,” Sproat said. “This center could serve as a test bed for developing these kinds of ideas across the board.”

This article is part of the Fall 2006 Synergy Issue, a publication of the Communications Office of the Beckman Institute.