Automatic training of phoneme dictionary based on mutual information criterion

Shigeki Okawa, Tetsunori Kobayashi, Katsuhiko Shirai

Research output: Contribution to journalConference articlepeer-review

Abstract

Proposes an automatic training mechanism for phoneme recognition using labelless speech data under the condition that only its orthographical phonemic symbol sequence is given. For the purpose of obtaining better recognition performance the authors attempt to realize an automatic labeling procedure based on a phoneme classification method by mutual information criterion. By iterative training of a phoneme dictionary for a large amount of speech data, one can investigate the performance and convergence properties of the dictionary. From experimental results, the percent correct of the labeling is over 98% after three iterations, and for the phoneme recognition, a very high accuracy is also obtained.

Original languageEnglish
Article number389310
Pages (from-to)I241-I244
JournalICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Volume1
DOIs
Publication statusPublished - 1994
EventProceedings of the 1994 IEEE International Conference on Acoustics, Speech and Signal Processing. Part 2 (of 6) - Adelaide, Aust
Duration: 1994 Apr 191994 Apr 22

ASJC Scopus subject areas

  • Software
  • Signal Processing
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Automatic training of phoneme dictionary based on mutual information criterion'. Together they form a unique fingerprint.

Cite this