Representing the Prosodic Context of Words using Gaussian Mixture Models

Speech Prosody, 2012

Shreyas A. Karkhedkar, Nigel G. Ward
Department of Computer Science, University of Texas at El Paso

Abstract: Different words tend to appear in different prosodic contexts For each word, we would like to be able to characterize its typ ical prosodic contexts of appearance, and, for applications pur poses, we would like to do this using probability distributions This paper describes the phenomena and the need, and explain a way to use Gaussian Mixture Models for modeling prosody.

Full Paper


Nigel Ward's Publications