SELECTION OF WORD-CANDIDATES BASED ON VECTOR QUANTIZATION FOR SPEAKER INDEPENDENT WORD RECOGNITION.

Katsuhiko Shirai, Koji Sano, Kazunori Mano

Research output: Contribution to journalArticle

Abstract

A method to select word-candidates for speaker independent word recognition is described. LEP cepstral coefficients are used as feature vectors, and vector quantization method is used to quantize and compress the features in each analysis frame to obtain speaker independence. Experiments were carried out for 100 city names uttered by 10 speakers with good results.

Original languageEnglish
Pages (from-to)68-75, 116
JournalBulletin of Centre for Informatics (Waseda University)
Volume3
Publication statusPublished - 1986 Mar 1
Externally publishedYes

ASJC Scopus subject areas

  • Engineering(all)

Fingerprint Dive into the research topics of 'SELECTION OF WORD-CANDIDATES BASED ON VECTOR QUANTIZATION FOR SPEAKER INDEPENDENT WORD RECOGNITION.'. Together they form a unique fingerprint.

  • Cite this