Computational Audio Lab

Our main research focus is on developing novel machine learning algorithms. We have a special emphasis on audio data. The lab is led by Prof. Paris Smaragdis, and there are students both from ECE and CS departments of University of Illinois at Urbana-Champaign (UIUC).


News

(4/28/2016)
Congratulations on Paris for being selected as a C.W. Gear Outstanding Junior Faculty.

(4/28/2016)
Minje was selected as an Outstanding Teaching Assistant for his TAship in Fall 2015, for the course "Machine Learning for Signal Processing".

(4/21/2016)
Minje successfully finished my PhD study! Here's the dissertation.

(12/21/2015)
Minje's paper was accepted for publication in ICASSP 2016:
"Efficient Neighborhood-Based Topic Modeling for Collaborative Audio Enhancement on Massive Crowdsourced Recordings"

(12/18/2015)
Congratulations on Paris for receiving NSF CAREER Award! (news article in DCS)

(10/23/2015)
Congratulations on Paris for being selected as the awardee of this year's C.W. Gear Outstanding Junior Faculty Award.

(7/15/2015)
Minje's paper was nominated for the best student paper on audio signal processing in the LVA/ICA conference:
"Adaptive Denoising Autoencoders: A Fine-tuning Scheme to Learn from Test Mixtures"

(6/29/2015)
Johannes and Cem have their papers accepted for publication in WASPAA this year:
"Directional NMF for Joint Source Separation and Localization"
"Method of Moments Learning for Left to Right Hidden Markov Models"

(5/22/2015)
Minje has an LVA/ICA paper accepted this year:
"Adaptive Denoising Autoencoders: A Fine-tuning Scheme to Learn from Unknown Sources"

(4/10/2015)
Congratulations on Paris for being selected as one of the Engineering Council Outstanding Advisors for two years straight, who represent the top 10% of engineering advisors selected by engineering students.

(1/14/2015)
Our paper, entitled "Efficient Manifold Preserving Audio Source Separation Using Locality Sensitive Hashing," was accepted for publication in ICASSP 2015.

(12/16/2014)
Congratulations on Paris for being named as a Fellow of IEEE.

(12/15/2014)
Minje's team was selected as one of the finalists for the Qualcomm Innovation Fellowship 2015.

(9/9/2014)
Our paper, entitled "Spectral Learning of Mixture of Hidden Markov Models," was accepted for publication in Neural Information Processing Systems 2014 (NIPS).

(7/7/2014)
Our paper, entitled "Multiple Speaker Tracking with the Factorial von Mises-Fisher Filter," was accepted for publication in IEEE Workshop on Machine Learning for Signal Processing (MLSP).

(6/27/2014)
Our paper, entitled "Mixture of Local Dictionaries for Unsupervised Speech Enhancement," was accepted for publication in the IEEE Signal Processing Letters.

(6/27/2014)
Our paper, entitled "Singing-Voice Separation From Monaural Recordings Using Deep Recurrent Neural Networks," was accepted for publication in the International Society for Music Information Retrieval Conference (ISMIR 2014).

(6/22/2014)
Our paper, Robust Interchannel Phase Difference Modeling with Wrapped Regression Splines, was presented in IEEE Sensor Array and Multichannel Signal Processing Workshop (SAM).

(6/10/2014)
Our paper, "Experiments on Deep Learning for Speech Denoising" was accepted for publication in Interspeech 2014.

(4/28/2014)
Congratulations on Paris for being selected as one of the Engineering Council Outstanding Advisors, who represent the top 10% of engineering advisors selected by engineering students.

(4/15/2014)
Minje won the Starkey Signal Processing Research Student Grant for the paper, "Deep Learning for Monaural Speech Separation (ICASSP 2014)."

(2/3/2014)
Congratulations on the acceptance of our papers,
"Phase and Level Difference Fusion for Robust Multichannel Source Separation"
and "Deep Learning for Monaural Speech Separation" in ICASSP 2014.





  • Contact:
    Siebel Center for Computer Science
    201 N. Goodwin Ave.
    Urbana, IL, 61801, USA
    Room #3231 (Paris)
    Room #3332 (Students)
    Map