Grapheme-to-Phoneme Conversion using Conditional Random Fields - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2011

Grapheme-to-Phoneme Conversion using Conditional Random Fields

Irina Illina
Dominique Fohr
Denis Jouvet

Abstract

We propose an approach to grapheme-to-phoneme conversion based on a probabilistic method: Conditional Random Fields (CRF). CRF give a long term prediction, assume relaxed state independence condition. Moreover, we propose an algorithm to one-to-one letter to phoneme alignment needed for CRF training. This alignment is based on discrete HMM. The proposed system is validated on two pronunciation dictionaries. Different CRF features are studied: POS-tag, context size, unigram versus bigram. Our approach compares favorably with the performance of the state-of-the-art Joint-Multigram Models for the quality of the pronunciations, but provides better recall and precision measures for multiple pronunciation variants generation.
No file

Dates and versions

inria-00614981 , version 1 (17-08-2011)

Identifiers

  • HAL Id : inria-00614981 , version 1

Cite

Irina Illina, Dominique Fohr, Denis Jouvet. Grapheme-to-Phoneme Conversion using Conditional Random Fields. 12th Annual Conference of the International Speech Communication Association - Interspeech 2011, International Speech Communication Association (ISCA) et The Italian Regional SIG - AISV (Italian Speech Communication Association), Aug 2011, Florence, Italy. ⟨inria-00614981⟩
270 View
0 Download

Share

Gmail Facebook Twitter LinkedIn More