Mark J. Harvilla

Professional Research Interests

Audio declipping, noise reduction, digital signal processing, automatic speech recognition (ASR), robust feature extraction for ASR, machine learning.

Education

Ph.D. in Electrical and Computer Engineering, Carnegie Mellon University, completed October 2014.

Advisor: Prof. Richard Stern

Thesis: "Compensation for Nonlinear Distortion in Noise for Robust Speech Recognition" [pdf] [slides]

M.S. in Electrical and Computer Engineering, Carnegie Mellon University, May 2013.

B.S. in Electrical and Computer Engineering, University of Pittsburgh, August 2010.

Employment

Chief Engineer, Oben @ Idealab, Pasadena, California, January 2015 - present. [company] [our iPhone apps]

Speech Engineer, Voci Technologies, Pittsburgh, Pennsylvania, June 2013 - December 2014. [company]

Publications

Mark J. Harvilla and Richard M. Stern (2015), “Robust Parameter Estimation for Audio Declipping in Noise,” In the proceedings of the 16th Annual Conference of the International Speech Communication Association (INTERSPEECH 2015), Dresden, Germany, 6-10 September. [pdf] [poster] [software]

Mark J. Harvilla and Richard M. Stern (2015), “Efficient Audio Declipping using Regularized Least Squares,” In the proceedings of the 2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Brisbane, Australia, 19-24 April. [pdf] [poster] [software]

Mark J. Harvilla and Richard M. Stern (2014), "Least Squares Signal Declipping for Robust Speech Recognition," In the proceedings of the 15th Annual Conference of the International Speech Communication Association (INTERSPEECH 2014), Singapore, 14-18 September. [pdf] [poster] [software]

Mark J. Harvilla and Richard M. Stern (2012), "Histogram-based Subband Power Warping and Spectral Averaging for Robust Speech Recognition under Matched and Multistyle Training," In the proceedings of the 2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Kyoto, Japan, 25-30 March. [pdf] [poster] [software]

Sourish Chaudhuri, Mark Harvilla, and Bhiksha Raj (2011), "Unsupervised Learning of Acoustic Unit Descriptors for Audio Content Representation and Classification," In the proceedings of the 12th Annual Conference of the International Speech Communication Association (INTERSPEECH 2011), Florence, 28-31 August. [pdf]

Guangyu Xia, Dawen Liang, Roger B. Dannenberg, and Mark J. Harvilla (2011), "Segmentation, Clustering, and Display in a Personal Audio Database for Musicians," In the proceedings of the 12th International Society for Music Information Retrieval Conference, Miami, USA, 24-28 October. [pdf]

Miscellaneous

My CV is available here. This is my Google Scholar page. I also play music, which you should listen to if you have a chance. I can be reached by e-mail with any comments or questions.