Prof. Paris Smaragdis of University of Illinois at Urbana-Champaign

Enabling machines to perceive the world using various modalities is one of the holy grails of artificial intelligence. In this talk I will present some research on creating machines that do as such by listening. I will discuss some of the unique difficulties in this field and present a thread of research which spans a range of computational disciplines relating to signal processing, machine learning and cryptography. This research will be introduced in the context of classic audio problems such as time/frequency analysis, music transcription, source separation, recognition in mixtures and more. I’ll show how this work generalizes and finds applications to other domains, what its practical implications are, and what it takes to move it from the whiteboard to the real-world.

Biography:

Paris Smaragdis (F) is faculty at the Computer Science and Electrical and Computer Engineering departments at the University of Illinois at Urbana-Champaign, and a Senior Research Scientist at Adobe Research. He completed his Masters (1997), Ph.D. (2001) and postdoctoral studies (2002) at the Machine Listening Group, MIT Media Lab. He was previously a research scientist at Mitsubishi Electric Research (MERL).

Prof. Smaragdis was selected by MIT’s Technology Review as one of the year’s top young technology innovators (TR35) for his work on machine listening, in 2006. In 2015, he was elevated to IEEE Fellow “for contributions in audio source separation and audio processing”. He has been elected as a full member, Acoustical Society of America (2008), and recipient of the C.W. Gear Outstanding Junior Faculty Award (2015), an NSF CAREER grant, and multiple teaching awards at the University of Illinois.

Prof. Smaragdis was Chair, IEEE Machine Learning for Signal Processing Technical Committee (2013-2014); Chair, LVA/ICA conference steering committee (2012-2015); Member, IEEE Machine Learning for Signal Processing Technical Committee (2010-2015); Member, Audio and Acoustic Signal Processing Technical Committee (2011-present); Associate Editor, IEEE Signal Processing Letters (2012-present); and Area Editor, IEEE Transactions for Signal Processing (2015-present). Prof. Smaragdis was an organizer of the GLOBALSIP Symposium on Machine Learning for Speech Processing (2014); General Co-Chair, Machine Learning for Signal Processing Workshop (2014); Technical Chair, IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (2011); and an organizer of multiple special sessions and tutorials at international conferences.

Prof. Smaragdis’ research is on applications of machine learning techniques on signal processing problems, especially as they apply to the analysis of sound mixtures. He has more than 120 publications in the areas of audio signal processing and machine learning, and holds 59 patents internationally.