Correlation modeling of MLLR transform biases for rapid HMM adaptation to new speakers

Abstract

This paper concerns rapid adaptation of hidden Markov model (HMM) based speech recognizers to a new speaker, when only few speech samples (one minute or less) are available from the new speaker. A widely used family of adaptation algorithms defines adaptation as a linearly constrained reestimation of the HMM Gaussians. With few speech data, tight… (More)
DOI: 10.1109/ICASSP.1999.759784

Topics

5 Figures and Tables

Cite this paper

@inproceedings{Bocchieri1999CorrelationMO, title={Correlation modeling of MLLR transform biases for rapid HMM adaptation to new speakers}, author={Enrico Bocchieri and Vassilios Digalakis and Adrian Corduneanu and Constantinos Boulis}, booktitle={ICASSP}, year={1999} }