Siddharth Mathur


E-mail: s.mathur [at] ieee.org

I graduated from the Electrical and Computer Engineering Department with an MS degree in December 2003. Since then, I have been working at a startup near Boston, Massachusetts: Nellymoser, Inc. This page documents my work with Dr. Brad Story at the Speech Acoustics Lab (August 2001 to December 2003). Please sends bouquets or brickbats to the e-mail address above, both are equally welcome!

Publications

• Mathur S., Story B.H., "Vocal tract modeling: Implementation of continuous length variations in a half-sample delay Kelly-Lochbaum model" [ps.gz] [PDF] [BibTeX]
  Proc. IEEE Intl. Symposium on Signal Processing and Information Technology (ISSPIT'03), Darmstadt, Germany, December 2003.

•  Mathur S., Story B.H., Rodriguez J.J., "Vocal-Tract Modeling: Fractional Elongation of Segment Lengths in a Waveguide Model With Half-Sample Delays"  [PDF] . IEEE Tran. Audio Speech and Language Processing, Vol. 14, No. 5, September 2006.
  

          MATLAB code for the above.

• Schwa vowel area function MAT file
• Basic digital waveguide model of tract [.m]
  

Reports

• MS Thesis, "Variable-length vocal tract modeling for speech synthesis"
  Submitted to the Electrical and Computer Engineering Department, University of Arizona. December 2003.

Research

My research at the Speech Acoustics Lab was related to articulatory speech synthesis, or physiologically-derived models of the human voice-production system. Specifically, I investigated ways to simulate continuous length changes in a digital waveguide model of the vocal tract. These enhancements allow speech scientists and engineers to map physiological data into computer model parameters with increased accuracy. The work leveraged previous ideas on the use of fractional delay filters and digital waveguides for vocal tract modeling. More details can be found in my MS thesis and the publications listed above.


Go to:
[SAL homepage]

Updated: January 22, 2006
This page is best viewed with any browser.