Speaker independent phonetic transcription of fluent speech for large vocabulary speech recognition

S. E. Levinson, M. Y. Liberman, A. Ljolje, L. G. Miller

Research output: Contribution to journalConference articlepeer-review

Abstract

Results are presented of experiments on speaker-independent phonetic transcription of fluent speech. The acoustic-phonetic model is a 38,505-parameter continuously variable duration hidden Markov model which allows real-time phonetic transcription to be performed by means of a modified Viterbi algorithm. The model was trained on 3020 sentences from the TIMIT database. Testing was performed on the remaining 180 sentences. In a test without lexical or syntactic constraints, the authors obtained 52% correct phonetic transcription with 12% insertions. The design of a system for recognition of fluent speech based on technique for phonetic transcription is described.

Original languageEnglish (US)
Pages (from-to)441-444
Number of pages4
JournalICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Volume1
StatePublished - Dec 1 1989
Externally publishedYes
Event1989 International Conference on Acoustics, Speech, and Signal Processing - Glasgow, Scotland
Duration: May 23 1989May 26 1989

ASJC Scopus subject areas

  • Software
  • Signal Processing
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Speaker independent phonetic transcription of fluent speech for large vocabulary speech recognition'. Together they form a unique fingerprint.

Cite this