ISCA Archive Interspeech 2008
ISCA Archive Interspeech 2008

An evaluation of non-standard features for grapheme-to-phoneme conversion

Gabriel Webster, Norbert Braunschweiler

Machine learning methods for grapheme-to-phoneme (G2P) conversion are popular, but the features used in the literature are most often simply a window of context letters, despite the availability of other features. In this paper, a set of features beyond the sevenletter window, termed non-standard features, are systematically evaluated for American English, using decision trees. The results show that adding non-standard features to a seven-letter window gives clear improvements for English, with the most important features being the previous three phone sequences predicted, an initial prediction of lexical stress location, and a window of vowel letters around the current letter.


doi: 10.21437/Interspeech.2008-180

Cite as: Webster, G., Braunschweiler, N. (2008) An evaluation of non-standard features for grapheme-to-phoneme conversion. Proc. Interspeech 2008, 1845-1848, doi: 10.21437/Interspeech.2008-180

@inproceedings{webster08_interspeech,
  author={Gabriel Webster and Norbert Braunschweiler},
  title={{An evaluation of non-standard features for grapheme-to-phoneme conversion}},
  year=2008,
  booktitle={Proc. Interspeech 2008},
  pages={1845--1848},
  doi={10.21437/Interspeech.2008-180}
}