In a strictly Bayesian approach, prior parameters are assumed known, based on common or subjective knowledge. But a practical solution for maximum a posteriori adaptation methods is to adopt an empirical Bayesian approach, where the prior parameters are estimated directly from training speech data itself. So there is a problem of mismatches between training and testing conditions in the use of prior parameters. We proposed a prior parameter transformation (PPT) adaptation approach that transforms the prior parameters to be more representative of the new speaker. In this paper we extend it to unsupervised mode. For easily confused speech units, different transformation matrices are applied to make them distinct. Initial experiments show that the PPT algorithm can get much improvement for a small amount of adaptation data even in the unsupervised mode.
Cite as: Li, G., Du, L., Hou, Z. (2000) Prior parameter transformation for unsupervised speaker adaptation. Proc. 6th International Conference on Spoken Language Processing (ICSLP 2000), vol. 3, 698-701, doi: 10.21437/ICSLP.2000-631
@inproceedings{li00j_icslp,
author={Guoqiang Li and Limin Du and Ziqiang Hou},
title={{Prior parameter transformation for unsupervised speaker adaptation}},
year=2000,
booktitle={Proc. 6th International Conference on Spoken Language Processing (ICSLP 2000)},
pages={vol. 3, 698-701},
doi={10.21437/ICSLP.2000-631}
}