ISCA Archive Interspeech 2008
ISCA Archive Interspeech 2008

Lexicon expansion using pronunciation variations extracted on the basis of speaker-related deviation in recognition error statistics

Yoshifumi Onishi

We propose a novel method for lexicon expansion using pronunciation variations extracted on the basis of speaker-related deviations in ASR error statistics. Two types of pronunciation variations were extracted: common pronunciation variations found with most speakers, and speaker-related pronunciation variations, identified on the basis of recognition error elements weighted by idf and tf-idf measures. Experimental results for CSJ show that entries added to the lexicon from speaker-related pronunciation variations were more effective than those generated on the basis of common pronunciation variations, some of which were superfluous.


doi: 10.21437/Interspeech.2008-497

Cite as: Onishi, Y. (2008) Lexicon expansion using pronunciation variations extracted on the basis of speaker-related deviation in recognition error statistics. Proc. Interspeech 2008, 1809-1812, doi: 10.21437/Interspeech.2008-497

@inproceedings{onishi08_interspeech,
  author={Yoshifumi Onishi},
  title={{Lexicon expansion using pronunciation variations extracted on the basis of speaker-related deviation in recognition error statistics}},
  year=2008,
  booktitle={Proc. Interspeech 2008},
  pages={1809--1812},
  doi={10.21437/Interspeech.2008-497}
}