Modeling acoustic transitions in speech by state-interpolation hidden Markov models

Author

Deng, Li ; Kenny, Patrick ; Lennig, Matthew ; Mermelstein, Paul

Author_Institution

INRS-Telecommun., Montreal, Que., Canada

Volume

40

Issue

2

fYear

1992

fDate

2/1/1992 12:00:00 AM

Firstpage

265

Lastpage

271

Abstract

The authors present a new type of hidden Markov model (HMM) for vowel-to-consonant (VC) and consonant-to-vowel (CV) transitions based on the locus theory of speech perception. The parameters of the model can be trained automatically using the Baum-Welch algorithm and the training procedure does not require that instances of all possible CV and VC pairs be present. When incorporated into an isolated word recognizer with a 75000 word vocabulary it leads to the modest improvement in recognition rates. The authors give recognition results for the state interpolation HMM and compare them to those obtained by standard context-independent HMMs and generalized triphone models

Keywords

Markov processes; interpolation; speech analysis and processing; speech recognition; Baum-Welch algorithm; HMM; acoustic transitions; consonant-to-vowel transitions; isolated word recognizer; locus theory; speech perception; speech recognition; state-interpolation hidden Markov models; training procedure; vowel-to-consonant transitions; Context modeling; Costs; Helium; Hidden Markov models; Interpolation; Speech recognition; Text recognition; Training data; Virtual colonoscopy; Vocabulary;

fLanguage

English

Journal_Title

Signal Processing, IEEE Transactions on

Publisher

ieee

ISSN

1053-587X

Type

jour

DOI

10.1109/78.124937

Filename

124937