Detailed pronunciation variant modeling for speech transcription - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2010

Detailed pronunciation variant modeling for speech transcription

Denis Jouvet
Dominique Fohr
Irina Illina

Abstract

Modeling pronunciation variants is an important topic for automatic speech recognition. This paper investigates the pronunciation modeling at the lexical level, and presents a detailed modeling of the probabilities of the pronunciation variants. The approach is evaluated on the French ESTER2 corpus, and a significant word error rate reduction is achieved through the use of context and speaking rate dependent modeling of these pronunciation probabilities. A rule-based approach makes it possible to derive a priori probabilities for the pronunciation of words that are not present in the training corpus, and a MAP estimation process yields reliable estimates of the pronunciation variant probabilities.
No file

Dates and versions

inria-00528225 , version 1 (21-10-2010)

Identifiers

  • HAL Id : inria-00528225 , version 1

Cite

Denis Jouvet, Dominique Fohr, Irina Illina. Detailed pronunciation variant modeling for speech transcription. INTERSPEECH, ISCA, Sep 2010, Makuhari, Japan. ⟨inria-00528225⟩
133 View
0 Download

Share

Gmail Facebook Twitter LinkedIn More