METHOD OF USING LINGUISTIC INFORMATION FOR AUTOMATIC SPOKEN WORD RECOGNITION.

Yoshihiro Niitsu, Jouji Miwa, Shozo Makino, Ken'iti Kido

Research output: Contribution to journalArticle

Abstract

The likelihood between phonemes is given by a confusion matrix based on the result of phoneme recognition and a method of matching with a word dictionary is proposed using dynamic programming with consideration of the probabilities of phoneme addition and omission. Using the proposed method, experiments in spoken word recognition are performed for a number of speakers and the effectiveness of the method is examined. The reliability of the confusion matrix is examined and it is shown that the proposed method is also effective for unspecified speakers. The results of recognition experiments for 25 male adults (including 10 unspecified speakers) were 84. 0, 95. 0 and 97. 2%, respectively, for 166, 51 and 20 city names.

Original languageEnglish
Pages (from-to)17-26
Number of pages10
JournalSystems, computers, controls
Volume10
Issue number1
Publication statusPublished - 1979 Jan 1

ASJC Scopus subject areas

  • Engineering(all)

Fingerprint Dive into the research topics of 'METHOD OF USING LINGUISTIC INFORMATION FOR AUTOMATIC SPOKEN WORD RECOGNITION.'. Together they form a unique fingerprint.

  • Cite this