DocumentCode :
1912352
Title :
Improved acoustic modeling with the SPHINX speech recognition system
Author :
Huang, X.D. ; Lee, K.F. ; Hon, H.W. ; Hwang, M.Y.
Author_Institution :
Sch. of Comput. Sci., Carnegie Mellon Univ., Pittsburgh, PA, USA
fYear :
1991
fDate :
14-17 Apr 1991
Firstpage :
345
Abstract :
The authors report recent efforts to further improve the performance of the SPHINX system for speaker-independent continuous speech recognition. They adhere to the basic architecture of the SPHINX system and use the DARPA resource management task and training corpus. The improvements are evaluated on the 600 sentences that comprise the DARPA February and October 1989 test sets. Several techniques that substantially reduced SPHINX´s error rate are presented. These techniques include dynamic features, semicontinuous hidden Markov models, speaker clustering, and the shared distribution modeling. The error rate of the baseline system was reduced by 45%
Keywords :
speech recognition; DARPA resource management task and training corpus; SPHINX speech recognition system; dynamic features; error rate reduction; improved acoustic modelling; semicontinuous hidden Markov models; shared distribution modeling; speaker clustering; speaker-independent continuous speech recognition; Cepstrum; Computer architecture; Computer science; Error analysis; Hidden Markov models; Linear predictive coding; Loudspeakers; Resource management; Smoothing methods; Speech recognition;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Acoustics, Speech, and Signal Processing, 1991. ICASSP-91., 1991 International Conference on
Conference_Location :
Toronto, Ont.
ISSN :
1520-6149
Print_ISBN :
0-7803-0003-3
Type :
conf
DOI :
10.1109/ICASSP.1991.150347
Filename :
150347
Link To Document :
بازگشت