DocumentCode :
336792
Title :
LSP weighting functions based on spectral sensitivity and mel-frequency warping for speech recognition in digital communication
Author :
Cho, Seung Ho ; Kim, Hong Kook ; Lee, Hwang Soo
Author_Institution :
Dept. of Electr. Eng., Korea Adv. Inst. of Sci. & Technol., Seoul, South Korea
Volume :
1
fYear :
1999
fDate :
15-19 Mar 1999
Firstpage :
401
Abstract :
In digital communication networks, a speech recognition system extracts feature parameters after reconstructing speech signals. In this paper, we consider a useful approach of incorporating speech coding parameters into a speech recognizer. Most speech coders employ line spectrum pairs (LSPs) to represent spectral parameters. We introduce weighted distance measures to improve the recognition performance of an LSP-based speech recognizer. Experiments on speaker-independent connected-digit recognition showed that weighted distance measures provide better recognition accuracy than unweighted distance measures do. Compared with a conventional method employing mel-frequency cepstral coefficients, the proposed method achieved higher performance in terms of a recognition accuracy
Keywords :
digital communication; spectral analysis; speech coding; speech recognition; telecommunication networks; LSP weighting functions; LSP-based speech recognizer; digital communication networks; experiments; feature parameters extraction; line spectrum pairs; mel-frequency cepstral coefficients; mel-frequency warping; recognition accuracy; recognition performance; speaker-independent connected-digit recognition; spectral parameters; spectral sensitivity; speech coders; speech coding parameters; speech recognition; speech signal reconstruction; unweighted distance measures; weighted distance measures; Degradation; Digital communication; Digital signal processing; Electronic mail; Euclidean distance; Feature extraction; Speech coding; Speech recognition; Telecommunications; Weight measurement;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Acoustics, Speech, and Signal Processing, 1999. Proceedings., 1999 IEEE International Conference on
Conference_Location :
Phoenix, AZ
ISSN :
1520-6149
Print_ISBN :
0-7803-5041-3
Type :
conf
DOI :
10.1109/ICASSP.1999.758147
Filename :
758147
Link To Document :
بازگشت