ISCA Archive Interspeech 2009
ISCA Archive Interspeech 2009

Grapheme to phoneme conversion using an SMT system

Antoine Laurent, Paul Deléglise, Sylvain Meignier

This paper presents an automatic grapheme to phoneme conversion system that uses statistical machine translation techniques provided by the Moses Toolkit. The generated word pronunciations are employed in the dictionary of an automatic speech recognition system and evaluated using the ESTER 2 French broadcast news corpus. Grapheme to phoneme conversion based on Moses is compared to two other methods: G2P, and a dictionary look-up method supplemented by a rule-based tool for phonetic transcriptions of words unavailable in the dictionary. Moses gives better results than G2P, and have performance comparable to the dictionary look-up strategy.


doi: 10.21437/Interspeech.2009-243

Cite as: Laurent, A., Deléglise, P., Meignier, S. (2009) Grapheme to phoneme conversion using an SMT system. Proc. Interspeech 2009, 708-711, doi: 10.21437/Interspeech.2009-243

@inproceedings{laurent09_interspeech,
  author={Antoine Laurent and Paul Deléglise and Sylvain Meignier},
  title={{Grapheme to phoneme conversion using an SMT system}},
  year=2009,
  booktitle={Proc. Interspeech 2009},
  pages={708--711},
  doi={10.21437/Interspeech.2009-243}
}