Title :
On-line adaptation of the SCHMM parameters based on the segmental quasi-Bayes learning for speech recognition
Author :
Huo, Qiang ; Chan, Chorkin ; Lee, Chin-Hui
Author_Institution :
Dept. of Comput. Sci., Hong Kong Univ., Hong Kong
fDate :
3/1/1996 12:00:00 AM
Abstract :
On-line quasi-Bayes adaptation of the mixture coefficients and mean vectors in semicontinuous hidden Markov model (SCHMM) is studied. The viability of the proposed algorithm is confirmed and the related practical issues are addressed in a specific application of on-line speaker adaptation using a 26-word English alphabet vocabulary
Keywords :
Bayes methods; adaptive systems; hidden Markov models; learning systems; speech recognition; English alphabet vocabulary; SCHMM; SCHMM parameters; mean vectors; mixture coefficients; online speaker adaptation; segmental quasiBayes learning; semicontinuous hidden Markov model; speech recognition; Acoustic testing; Acoustic transducers; Bayesian methods; Computer science; Hidden Markov models; Loudspeakers; Robustness; Speech recognition; System testing; Vocabulary;
Journal_Title :
Speech and Audio Processing, IEEE Transactions on
