DocumentCode :
696754
Title :
Maximum a posteriori linear regression for speaker adaptation with the prior of mean
Author :
Lin, Chih-Heng ; Wang, Wern-Jun
Author_Institution :
Chunghwa Telecommunication Laboratories, 12, lane 551, min-tsu rd. sec 5, yang-mei, Taoyuan, Taiwan 326
fYear :
2000
fDate :
4-8 Sept. 2000
Firstpage :
1
Lastpage :
4
Abstract :
An efficient method for speaker adaptation (SA) is proposed in this paper. Let the relationship between the mean parameters of adapted model and the mean parameters of the speaker independent (SI) model be represented by sets of linear transformations like that of maximum likelihood linear regression (MLLR) approach, we try to estimate the transformations by maximum a posteriori (MAP) criterion. The prior mean distribution is considered in the estimation. The experiments on Mandarin speech recognition show the proposed approach is superior to the MLLR approach when only little speech is available for speaker adaptation.
Keywords :
Adaptation models; Covariance matrices; Estimation; Hidden Markov models; Silicon; Speech; Speech recognition;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Signal Processing Conference, 2000 10th European
Conference_Location :
Tampere, Finland
Print_ISBN :
978-952-1504-43-3
Type :
conf
Filename :
7075375
Link To Document :
بازگشت