MVDR modeling of Speech and its application to Speech Recognition

Dr. Bhaskar D. Rao
University of California, San DiegoAbstract: In this talk, we present all-pole models based upon the minimum variance distortionless response (MVDR) spectrum for spectral modeling of speech. Although linear prediction (LP) is a popular method for obtaining all-pole model parameters, LP spectral envelopes overestimate and overemphasize the medium and high pitch voiced speech spectral powers, thereby featuring unwanted sharp contours, and do not improve in spectral envelope modeling performance as the filter order is increased. In contrast, the MVDR all-pole spectrum, which can be easily obtained from the LP coefficients, is shown to feature improved spectral envelope modeling as the filter order is increased. In particular, the high order MVDR spectrum models voiced speech spectra very well, particularly at the perceptually important harmonics, and features a smooth contoured envelope. Computationally, the MVDR spectrum can be based upon either conventional time domain correlation estimates or upon spectral samples, a task that is common in frequency domain speech coding. A robust feature extraction technique for continuous speech recognition based on MVDR modeling is discussed. We consider incorporating perceptual information in two ways: (1) after the MVDR power spectrum is computed and (2) directly during the MVDR spectrum estimation. Incorporating perceptual information directly into the spectrum estimation improves both robustness and computational efficiency significantly. We evaluate the technique on several recognition tasks and the results compare favorably with existing techniques.About the Speaker: Bhaskar D. Rao received the B.Tech. degree in electronics and electrical communication engineering from the Indian Institute of Technology, Kharagpur, India, in 1979, and the M.S. and Ph.D. degrees from the University of Southern California, Los Angeles, in 1981 and 1983, respectively. Since 1983, he has been with the University of California, San Diego, where he is currently a Professor in the Electrical and Computer Engineering department. He is a Fellow of IEEE. His interests are in the areas of digital signal processing, estimation theory, and optimization theory, with applications to digital communications, speech signal processing, and human-computer interactions.Host: Shri Narayanan (ext 06432)

Location: Ronald Tutor Hall of Engineering (RTH) - 211

Audiences: Everyone Is Invited

Contact: Talyia Veal