Discriminative Linear Transforms for Feature Normalization and Speaker Adaptation in HMM Estimation

Download: PDF, slides.

“Discriminative Linear Transforms for Feature Normalization and Speaker Adaptation in HMM Estimation” by S. Tsakalidis, V. Doumpiotis, and W. Byrne. In Proc. of the International Conference on Spoken Language Processing, (Denver, Colorado, USA), 2002.

Abstract

Linear transforms have been used extensively for training and adaptation of HMM-based ASR systems. Recently procedures have been developed for the estimation of linear transforms under the Maximum Mutual Information (MMI) criterion. In this paper we introduce discriminative training procedures that employ linear transforms for feature normalization and for speaker adaptive training. We integrate these discriminative linear transforms into MMI estimation of HMM parameters for improvement of large vocabulary conversational speech recognition systems.

Download: PDF, slides.

BibTeX entry:

@inproceedings{dllt_icslp02,
   author = {S. Tsakalidis and V. Doumpiotis and W. Byrne,},
   title = {Discriminative Linear Transforms for Feature Normalization and
	Speaker Adaptation in {HMM} Estimation},
   booktitle = {Proc. of the International Conference on Spoken Language
	Processing},
   pages = {(5 pages)},
   address = {Denver, Colorado, USA},
   year = {2002}
}

Back to Bill Byrne publications.