Automatic recognition of conversational speech tends to have higher word error rates (WER) than read speech. Improvements gained from unsupervised speaker adaptation methods like Maximum Likelihood Linear Regression (MLLR) [1] are reduced because of their sensitivity to recognition errors in the first pass. We show that a more detailed modeling of adaptation classes and the use of con- fidence measures improve the adaptation performance. We present experimental results on the VERBMOBIL task, a German conversational speech corpus.
C.J. Leggetter, P.C.Woodland: "Maximum Likelihood linear regression for speaker adaptation of continuous density hidden Markov models", Computer, Speech and Language, vol. 9, pp. 171-185, 1995.
Cite as: Pitz, M., Wessel, F., Ney, H. (2000) Improved MLLR speaker adaptation using confidence measures for conversational speech recognition. Proc. 6th International Conference on Spoken Language Processing (ICSLP 2000), vol. 4, 548-551, doi: 10.21437/ICSLP.2000-870
@inproceedings{pitz00_icslp,
author={Michael Pitz and Frank Wessel and Hermann Ney},
title={{Improved MLLR speaker adaptation using confidence measures for conversational speech recognition}},
year=2000,
booktitle={Proc. 6th International Conference on Spoken Language Processing (ICSLP 2000)},
pages={vol. 4, 548-551},
doi={10.21437/ICSLP.2000-870}
}