Title :
A model block-training method for HMM-based speech recognition systems
Author :
Gao, Yu-Qing ; Xie, Jin-Hui
Abstract :
A model block-training method for hidden Markov models (HMMs) is described. It combines several model estimations from several training sets, all of which are derived from utterances of the same word, into a new one. Although the recognition rate of the recognizer trained by the block-training method is lower than that of a recognizer trained by a batch-training method, the new one is much more flexible than the old one in building up the multispeaker or speaker-independent system in spite of the significant difference in spectral features. A computer simulation shows that the block-training procedure can converge to a local maximum point of likelihood function
Keywords :
Markov processes; learning systems; probability; spectral analysis; speech recognition; hidden Markov models; likelihood function; local maximum point; model block-training; speaker-independent system; speech recognition; utterances; Automation; Computer simulation; Heuristic algorithms; Hidden Markov models; Parameter estimation; Pattern recognition; Speech analysis; Speech processing; Speech recognition; Training data;
Conference_Titel :
Acoustics, Speech, and Signal Processing, 1990. ICASSP-90., 1990 International Conference on
Conference_Location :
Albuquerque, NM
DOI :
10.1109/ICASSP.1990.115769