Interdisciplinary Approaches for Advancing Articulatory Speech Theory and Synthesis

  • Sidney Fels University of British Columbia
  • Bryan Gick University of British Columbia

Abstract

For years, articulatory synthesis research has been largely overshadowed by frequency domain and concatenate sample-based speech synthesis techniques. While successful in some domains (e.g., voice-based databases), these techniques still cannot produce natural looking and sounding speech from text from an arbitrary speaker. Natural looking and sounding speech technology is one of the next major milestones in voice-based interaction for natural user interfaces. Through a team of interdisciplinary researchers, we have been steadily working towards creating the necessary platform to overcome basic problems in speech production and, we believe, represents the next major advance in speech synthesis technology. We will discuss our progress on articulatory speech synthesis using 3D biomechanical models of the vocal tract and the advances in understanding of speech production and synthesis produced from it.

Author Biographies

Sidney Fels, University of British Columbia
Professor, Department of Electrical and Computer Engineering
Bryan Gick, University of British Columbia
Professor and Head, Department of Linguistics
Published
2016-08-26
How to Cite
1.
Fels S, Gick B. Interdisciplinary Approaches for Advancing Articulatory Speech Theory and Synthesis. Canadian Acoustics [Internet]. 2016Aug.26 [cited 2019Dec.16];44(3). Available from: //jcaa.caa-aca.ca/index.php/jcaa/article/view/3018
Section
Proceedings of the Acoustics Week in Canada

Most read articles by the same author(s)