Investigations on joint-multigram models for grapheme-to-phoneme conversion

@inproceedings{Bisani2002InvestigationsOJ,
  title={Investigations on joint-multigram models for grapheme-to-phoneme conversion},
  author={Maximilian Bisani and Hermann Ney},
  booktitle={INTERSPEECH},
  year={2002}
}
We present a fully data-driven, language independent way of building a grapheme-to-phoneme converter. We apply the joint-multigram approach to the alignment problem and use standard language modelling techniques to model transcription probabilities. We study model parameters, training procedures and effects of corpus size in detail. Experiments were conducted on English and German pronunciation lexica. Our proposed training scheme performs better than previously published ones. Phoneme error… CONTINUE READING

Figures, Tables, Results, and Topics from this paper.

Key Quantitative Results

  • Phoneme error rates as low as 3.98% for English and 0.51% for German were achieved.

Citations

Publications citing this paper.
SHOWING 1-10 OF 69 CITATIONS, ESTIMATED 24% COVERAGE

Letter-to-Phoneme Conversion for a German Text-to-Speech System

VIEW 9 EXCERPTS
CITES METHODS & BACKGROUND
HIGHLY INFLUENCED

Grapheme-to-phoneme conversion methods for minority language conditions

  • 2012 International Conference on Speech Database and Assessments
  • 2012
VIEW 9 EXCERPTS
CITES METHODS & RESULTS
HIGHLY INFLUENCED

Coping with out-of-vocabulary words: Open versus huge vocabulary asr

  • 2009 IEEE International Conference on Acoustics, Speech and Signal Processing
  • 2009
VIEW 3 EXCERPTS
CITES METHODS & RESULTS
HIGHLY INFLUENCED

Using graphone models in automatic speech recognition

VIEW 3 EXCERPTS
CITES RESULTS, METHODS & BACKGROUND
HIGHLY INFLUENCED

Adapting grapheme-to-phoneme conversion for name recognition

  • 2007 IEEE Workshop on Automatic Speech Recognition & Understanding (ASRU)
  • 2007
VIEW 4 EXCERPTS
CITES METHODS & BACKGROUND
HIGHLY INFLUENCED

FILTER CITATIONS BY YEAR

2003
2018

CITATION STATISTICS

  • 7 Highly Influenced Citations

  • Averaged 1 Citations per year over the last 3 years

  • 100% Increase in citations per year in 2018 over 2017

References

Publications referenced by this paper.
SHOWING 1-3 OF 3 REFERENCES

Bielefelder Lexikon und Morphologie in V ERBMOBIL Phase II

K. Ehlebracht H. Lüngen, D. Gibbon, A. P. Q. Simões
  • Konferenz zur Verarbeitung natürlicher Sprache ( KONVENS )

Similar Papers

Loading similar papers…