A real time speaker-independent continuous speech recognition system

Author

Iwasaki, Tomohiro ; Nakajima, Kunio

Author_Institution

Comput. & Inf. Syst. Lab., Mitsubishi Electric Corp., Kamakura, Japan

fYear

1992

fDate

30 Aug-3 Sep 1992

Firstpage

663

Lastpage

666

Abstract

Reports on the speaker-independent continuous speech recognition system with speaker adaptation. This system is based on continuous HMM with mixture Gaussian distribution, and several fast algorithms are applied for real time calculation. To reduce HMM states, the system uses the mono-state context dependent model, called acoustic phonetic segment. The calculation of mixture Gaussian distributions is reduced, by varying the number of mixtures dynamically. The fast Viterbi calculation algorithm with duration control is used. The system has been successfully implemented as a man-machine interface of the plant control expert system, achieving sentence accuracy of 98.7%

Keywords

hidden Markov models; real-time systems; speech recognition equipment; HMM; acoustic phonetic segment; fast Viterbi calculation algorithm; hidden Markov models; man-machine interface; mixture Gaussian distribution; real time systems; speaker adaptation; speaker-independent continuous speech recognition system; Context modeling; Control systems; Expert systems; Gaussian distribution; Hidden Markov models; Loudspeakers; Real time systems; Speech recognition; User interfaces; Viterbi algorithm;

fLanguage

English

Publisher

ieee

Conference_Titel

Pattern Recognition, 1992. Vol.III. Conference C: Image, Speech and Signal Analysis, Proceedings., 11th IAPR International Conference on

Conference_Location

The Hague

Print_ISBN

0-8186-2920-7

Type

conf

DOI

10.1109/ICPR.1992.202073

Filename

202073