In this paper, a new approach for model adaptation, extended maximum a posterior linear regression (EMAPLR), is described and studied. EMAPLR is an extension of maximum a posterior linear regression (MAPLR) for transform based model adaptation. The proposed approach has a close form solution under the elliptic symmetric matrix variate priors, and it is effective in our speech recognition experiments. EMAPLR is based on a direct MAPLR solution of the transform imageWs without explicitly solving the transformation matrix W. This is fundamentally different from conventionalMAPLR and MLLR. Moreover, the proposed EMAPLR approach is incorporated with the structured prior evolution which significantly improves the algorithm efficiency and robustness. The structure of prior evolution in MAPLR is studied and it is shown that under the structured prior evolution, the priors in MAPLR follows a recursive formulation. Experimental results on WSJ (Spoke 3) non-native speaker adaptation task indicates that significant gain over MLLR and MAPLR can be obtained with same amount of adaptation data.
Cite as: Chou, W., Siohan, O., Myrvoll, T.A., Lee, C.-H. (2000) Extended maximum a posterior linear regression (EMAPLR) model adaptation for speech recognition. Proc. 6th International Conference on Spoken Language Processing (ICSLP 2000), vol. 4, 616-619, doi: 10.21437/ICSLP.2000-887
@inproceedings{chou00b_icslp,
author={Wu Chou and Olivier Siohan and Tor André Myrvoll and Chin-Hui Lee},
title={{Extended maximum a posterior linear regression (EMAPLR) model adaptation for speech recognition}},
year=2000,
booktitle={Proc. 6th International Conference on Spoken Language Processing (ICSLP 2000)},
pages={vol. 4, 616-619},
doi={10.21437/ICSLP.2000-887}
}