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

Download: PDF.

Discriminative Linear Transforms for Feature Normalization and Speaker Adaptation in HMM Estimation” by V. Doumpiotis, S. Tsakalidis, and W. Byrne. IEEE Transactions on Speech and Audio Processing, vol. 13, no. 3, May 2005, pp. 367-376 (10 pages).

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.

BibTeX entry:

@article{dlt_spatrans,
   author = {V. Doumpiotis and S. Tsakalidis and W. Byrne},
   title = {Discriminative Linear Transforms for Feature Normalization and
	Speaker Adaptation in {HMM} Estimation},
   journal = {IEEE Transactions on Speech and Audio Processing},
   volume = {13},
   number = {3},
   pages = {367--376 (10 pages)},
   month = may,
   year = {2005},
   url = {http://dx.doi.org/10.1109/TSA.2005.845806}
}

Back to Bill Byrne publications.