Comparison of Speaker Adaptation Methods as Feature Extraction for SVM-Based Speaker Recognition

Abstract

In the last years the speaker recognition field has made extensive use of speaker adaptation techniques. Adaptation allows speaker model parameters to be estimated using less speech data than needed for maximum-likelihood (ML) training. The maximum <i>a posteriori</i> (MAP) and maximum-likelihood linear regression (MLLR) techniques have typically been used… (More)
DOI: 10.1109/TASL.2009.2034187

Topics

9 Figures and Tables

Cite this paper

@article{Ferras2010ComparisonOS, title={Comparison of Speaker Adaptation Methods as Feature Extraction for SVM-Based Speaker Recognition}, author={Marc Ferras and Cheung-Chi Leung and Claude Barras and Jean-Luc Gauvain}, journal={IEEE Transactions on Audio, Speech, and Language Processing}, year={2010}, volume={18}, pages={1366-1378} }