7th International Conference on Spoken Language Processing

September 16-20, 2002
Denver, Colorado, USA

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

Stavros Tsakalidis, Vlasios Doumpiotis, William Byrne

Johns Hopkins University, USA

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.


Full Paper

Bibliographic reference.  Tsakalidis, Stavros / Doumpiotis, Vlasios / Byrne, William (2002): "Discriminative linear transforms for feature normalization and speaker adaptation in HMM estimation", In ICSLP-2002, 2585-2588.