Speaker clustering and transformation for speaker adaptation in large-vocabulary speech recognition systems
- 24 December 2002
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 2 (15206149) , 701-704
- https://doi.org/10.1109/icassp.1996.543217
Abstract
A speaker adaptation strategy is described that is based on finding a subset of speakers, from the training set, who are acoustically close to the test speaker, and using only the data from these speakers (rather than the complete training corpus) to re-estimate the system parameters. Further, a linear transformation is computed for every one of the selected training speakers to better map the training speaker's data to the test speaker's acoustic space. Finally, the system parameters (Gaussian means) are re-estimated specifically for the test speaker using the transformed data from the selected training speakers. Experiments showed that this scheme is capable of reducing the error rate by 10-15% with the use of as little as 3 sentences of adaptation data.Keywords
This publication has 6 references indexed in Scilit:
- Batch, incremental and instantaneous adaptation techniques for speech recognitionPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Experiments using data augmentation for speaker adaptationPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Performance of the IBM large vocabulary continuous speech recognition system on the ARPA Wall Street Journal taskPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Speaker adaptation using combined transformation and Bayesian methodsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Maximum likelihood linear regression for speaker adaptation of continuous density hidden Markov modelsComputer Speech & Language, 1995
- Maximum a posteriori estimation for multivariate Gaussian mixture observations of Markov chainsIEEE Transactions on Speech and Audio Processing, 1994