Duration modeling with expanded HMM applied to speech recognition

Author

Bonafonte, Antonio ; Vidal, Josep ; Nogueiras, Albino

Author_Institution

Univ. Politecnica de Catalunya, Barcelona, Spain

Volume

2

fYear

1996

fDate

3-6 Oct 1996

Firstpage

1097

Abstract

The occupancy of the HMM states is modeled by means of a Markov chain. A linear estimator is introduced to compute the probabilities of the Markov chain. The distribution function (DF) represents accurately the observed data. Representing the DF as a Markov chain allows the use of standard HMM recognizers. The increase of complexity is negligible in training and strongly limited during recognition. Experiments performed on acoustic-phonetic decoding shows how the phone recognition rate increases from 60.6 to 61.1. Furthermore, on a task of database inquires, where phones are used as subword units, the correct word rate increases from 88.2 to 88.4

Keywords

decoding; hidden Markov models; probability; speech recognition; HMM state occupancy; Markov chain; acoustic-phonetic decoding; complexity; correct word rate; database inquires; distribution functions; duration modeling; expanded HMM; linear estimator; phone recognition rate; probabilities; speech recognition; standard HMM recognizers; subword units; training; Computational complexity; Databases; Decoding; Distribution functions; Hidden Markov models; Parameter estimation; Probability density function; Speech recognition; State estimation; Training data;

fLanguage

English

Publisher

ieee

Conference_Titel

Spoken Language, 1996. ICSLP 96. Proceedings., Fourth International Conference on

Conference_Location

Philadelphia, PA

Print_ISBN

0-7803-3555-4

Type

conf

DOI

10.1109/ICSLP.1996.607797

Filename

607797