Fast search algorithm for VQ-based recognition of isolated words

Author

Chen, S.-H. ; Pan, J.S.

Author_Institution

Dept. of Commun. Eng., Nat. Chiao Tung Univ., Hsinchu, Taiwan

Volume

136

Issue

6

fYear

1989

fDate

12/1/1989 12:00:00 AM

Firstpage

391

Lastpage

396

Abstract

The authors present a fast search algorithm for vector quantisation (VQ)-based recognition of isolated words. It incorporates the property of high correlation between speech feature vectors of consecutive frames with the method of triangular inequality elimination to relieve the computational burden of vector-quantising the test feature vectors by full code-book search, and uses the extended partial distortion method to compress the incomplete matching computations of wildly mismatched words. Overall computational load can therefore be drastically reduced while the recognition performance of full search can be retained. Experimental results show that about 93% of multiplications and additions can be saved with a little increase of both comparisons and memory space.

Keywords

encoding; speech recognition; additions; consecutive frames; extended partial distortion method; fast search algorithm; full code-book search; high correlation; isolated word recognition; memory space; multiplications; speech feature vectors; speech recognition; test feature vectors; triangular inequality elimination; vector quantisation-based recognition; wildly mismatched words;

fLanguage

English

Journal_Title

Communications, Speech and Vision, IEE Proceedings I

Publisher

iet

ISSN

0956-3776

Type

jour

Filename

268934