Speech Recognition of lsolated Digits Using Simultaneous Generative Histogram

Yasuhisa HAYASHI  Akio OGIHARA  Kunio FUKUNAGA  

Publication
IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences   Vol.E76-A   No.12   pp.2052-2054
Publication Date: 1993/12/25
Online ISSN: 
DOI: 
Print ISSN: 0916-8508
Type of Manuscript: Special Section LETTER (Special Section of Letters Selected from the 1993 IEICE Fall Conference)
Category: 
Keyword: 
speech recognition,  hidden Markov model,  separate vector quantization,  simultaneous generative histogram,  

Full Text: PDF(214KB)
>>Buy this Article


Summary: 
We propose a recognition method for HMM using a simultaneous generative histogram. Proposed method uses the correlation between two features, which is expressed by a simultaneous generative histogram. Then output probabilities of integrated HMM are conditioned by the codeword of another feature. The proposed method is applied to isolated digit word recognition to confirm its validity.