Advertisement

Comparison of Grapheme and Phoneme Based Acoustic Modeling in LVCSR Task in Slovak

  • Michal Mirilovič
  • Jozef Juhár
  • Anton Čižmár
Conference paper
  • 839 Downloads
Part of the Lecture Notes in Computer Science book series (LNCS, volume 5398)

Abstract

Phonemes and allophones are the basic speech units for acoustic modeling in the majority of contemporary HMM based speech recognizers. Grapheme-based acoustic sub-word units were applied to multi-lingual and cross-lingual acoustic modeling in many tasks. Grapheme and phoneme based mono-, cross- and bilingual speech recognition of Czech and Slovak in the small and medium vocabulary task has been studied in our previous work. In this article we compare grapheme and phoneme based approach to acoustic modeling and model unit selection in large vocabulary continuous speech recognition (LVCSR) task in Slovak. The main goal of our experimental work is to investigate a possibility to select an optimal set of sub-word units for Slovak LVCSR system.

Keywords

Speech Recognition Acoustic Modeling Speech Recognition System Word Error Rate Phonetic Transcription 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Schukat-Talamazzini, E.G., Niemann, H., Eckert, W., Kuhn, T., Rieck, S.: Automatic speech recognition without phonemes. In: Proceeding of the Eurospeech, Berlin, September 22-25, pp. 129–132 (1993)Google Scholar
  2. 2.
    Magimai-Doss, M., Stephenson, T.A., Bourlard, H., Bengio, S.: Phoneme-grapheme based speech recognition system. In: Proceedings of 2003 IEEE Workshop on Automatic Speech Recognition and Understanding, St. Thomas, U.S. Virgin Islands, November 30 - December 4, pp. 94–98 (2003)Google Scholar
  3. 3.
    Magimai-Doss, M., Bengio, S., Bourlard, H.: Joint decoding for phoneme-grapheme continuous speech recognition. In: Proceedings of ICASSP, Quebec, Kanada, May 17-21, pp. 177–180 (2004)Google Scholar
  4. 4.
    Kanthak, S., Ney, H.: Multilingual acoustic modeling using graphemes. In: Proceeding of the Eurospeech, Geneva, Switzerland, September 1-4, pp. 1145–1148 (2003)Google Scholar
  5. 5.
    Killer, M., Stüker, S., Schultz, T.: Grapheme based speech recognition. In: Proceeding of the Eurospeech, Geneva, Switzerland, September 1-4, pp. 3141–3144 (2003)Google Scholar
  6. 6.
    Schultz, T.: Towards rapid language portability of speech processing systems. In: Proceedings of the Conference on Speech and Language Systems for Human Communication, SPLASH 2004, Delhi, India, November 17-19 (2004)Google Scholar
  7. 7.
    Rubagotti, E.: Is it possible to train a speech recognition system on text only? In: Interspeech 2006 - ICSLP, Stellenbosch, South Africa, April 9-11 (2006)Google Scholar
  8. 8.
    Le, V.B., Besacier, L.: Comparison of acoustic modeling techniques for vietnamese and khmer asr. In: Interspeech 2006 - ICSLP, Pittsburgh, USA, September 17-21, pp. 129–132 (2006)Google Scholar
  9. 9.
    Charoenpornsawat, P., Hewavitharana, S., Schultz, T.: Thai grapheme-based speech recognition. In: Proc. of the HLT-NAACL, New York City, USA, June 5-7, pp. 17–20 (2006)Google Scholar
  10. 10.
    Stüker, S., Schultz, T.: A grapheme based speech recognition system for Russian. In: Proceedings of SPECOM 2004, Petersburgh, Russia, September 20-22 (2004)Google Scholar
  11. 11.
    Kanthak, S., Ney, H.: Context-dependent acoustic modeling using graphemes for large vocabulary speech recognition. In: Proceeding of the ICASSP, Orlando, Florida, May 13-17, pp. 845–848 (2002)Google Scholar
  12. 12.
    Schillo, C., Fink, G.A., Kummert, F.: Grapheme based speech recognition for large vocabularies. In: Proceeding of the ICSLP, Beijing, China, October 16-20, pp. 584–587 (2000)Google Scholar
  13. 13.
    Lihan, S., Juhár, J., Čižmár, A.: Comparison of Slovak and Czech speech recognition based on grapheme and phoneme acoustic models. In: Interspeech 2006 - ICSLP, Pittsburgh, USA, September 17-21, pp. 149–152 (2006)Google Scholar
  14. 14.
    Mirilovič, M., Juhár, J., Čižmár, A.: Large vocabulary continuous speech recognition in slovak. In: Proc. Int. Conf. on Applied Electrical Engineering and Informatics - AEI 2008, Greece, September 8-11 (2008)Google Scholar
  15. 15.
    Lindberg, B., Johansen, F.T., Warakagoda, N., Lehtinen, G., Kačič, Z., Žgank, A., Elenius, K., Salvi, G.: A noise robust multilingual reference recogniser based on SpeechDat(II). In: Proc. ICSLP 2000, Beijing, China, October 16-20, vol. 3, pp. 370–373 (2000)Google Scholar
  16. 16.
    Šimková, M.: Slovak national corpus history and current situation. In: Insight into the Slovak and Czech Corpus Linguistics, Veda, Bratislava, pp. 151–159 (2006)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2009

Authors and Affiliations

  • Michal Mirilovič
    • 1
  • Jozef Juhár
    • 1
  • Anton Čižmár
    • 1
  1. 1.Department of Electronics and Multimedia CommunicationTechnical University of KošiceSlovakia

Personalised recommendations