An SVQ-HMM Training Method Using Simultaneous Generative Histogram

Yasuhisa HAYASHI  Satoshi KONDO  Nobuyuki TAKASU  Akio OGIHARA  Shojiro YONEDA  

Publication
IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences   Vol.E75-A   No.7   pp.905-907
Publication Date: 1992/07/25
Online ISSN: 
DOI: 
Print ISSN: 0916-8508
Type of Manuscript: Special Section LETTER (Special Section on the 1992 IEICE Spring Conference)
Category: 
Keyword: 
speech recognition,  separate vector quantization,  hidden Markov model,  simultaneous generative histogram,  

Full Text: PDF(212KB)>>
Buy this Article




Summary: 
This study proposes a new training method for hidden Markov model with separate vector quantization (SVQ-HMM) in speech recognition. The proposed method uses the correlation of two different kinds of features: cepstrum and delta-cepstrum. The correlation is used to decrease the number of reestimation for two features thus the total computation time for training models decreases. The proposed method is applied to Japanese language isolated dgit recognition.