9/2/2010
Written by
The University of Illinois department of computer science is pleased to welcome Professor Paris Smaragdis to the faculty. Prof. Smaragdis is a joint computer science and electrical and computer engineering faculty member. His work on signal processing, machine learning and statistics as they relate to artificial perception, and computational audition promises to open up exciting new avenues for collaboration at Illinois.
Named one of MIT Technology Review’s Top Young Technology Innovators in 2006, Prof. Smaragdis’ primary research interests lie at the intersection of machine learning and signal processing. His current work to create machines that can listen is pioneering new devices to improve music recordings, safety, and even televised broadcasts.
Dealing with superimposed signals is one of the most challenging problems in audio and speech processing today. Smaragdis’ work has discovered a new way to approach this problem and he’s created state-of-the-art methods for working with mixed signals. Smaradgis’ research explores the computational foundations for constucting systems that can understand sound the same way that humans do.
“Making computers that understand their world around them is an incredibly hard problem. Fortunately it is also fascinating,” says Smaragdis. “On the theoretical side my work involves creating new tools for processing and analyzing time-series. On the practical side this results in constructing actual machines with hearing abilities, such as TVs that can find when the football game gets interesting, stethoscopes that detect and analyze heartbeats, music players that automatically DJ for you and smart traffic lights that can hear accidents that happen in their intersection.”
Smaragdis is also interested in anything involving audio and computation and have been involved in the fields of computer music, audio synthesis algorithms and real-time performance.
"Illinois has a long track record of innovation in the area of computational perception," says Rob A. Rutenbar, Bliss Professor and Head of Computer Science. "Machine vision, machine reading and machine learning are all core parts of the CS discipline. Paris adds a whole new dimension to our department: machine listening. His work at the intersection of machine learning and signal processing is generating remarkable new methods for manipulating acoustic information in novel ways."
Adds Andreas Cangellaris, Van Valkenburg Professor and Head of Electrical and Computer Engineering, "Paris' work is relevant to numerous real-life applications, ranging from speech processing, to psychoacoustics, to security applications, to tools creation for artists."
Smaragdis was formerly a senior research scientist at Adobe Systems. He completed his graduate and postdoctoral studies at MIT, where he conducted research on computational perception. Prior to Adobe he was a research scientist at Mitsubishi Electric Research Labs, during which time he was selected by the MIT Technology Review as one of the top 35 young innovators of 2006.
Paris will be teaching CS 598 (Machine Learning for Signal Processing) in the Fall semester.