DocumentCode :
1540428
Title :
Speaker adaptation using learning vector quantisation
Author :
Koo, M.W. ; Un, C.K.
Author_Institution :
Dept. of Electr. Eng., Korea Adv. Inst. of Sci. & Technol., Seoul, South Korea
Volume :
26
Issue :
20
fYear :
1990
Firstpage :
1731
Lastpage :
1732
Abstract :
A codebook adaptation scheme using learning vector quantisation (LVQ) which has highly discriminant ability for speaker adaptation is presented. The recognition accuracy for new speakers can be improved over the VQ adaptation method based on the minimum distortion criterion using this scheme.
Keywords :
learning systems; speech recognition; codebook adaptation scheme; discriminant ability; learning vector quantisation; recognition accuracy; speaker adaptation;
fLanguage :
English
Journal_Title :
Electronics Letters
Publisher :
iet
ISSN :
0013-5194
Type :
jour
DOI :
10.1049/el:19901105
Filename :
58207
Link To Document :
بازگشت