Decision Tree-Based Context Dependent Sublexical Units for Continuous Speech Recognition of Basque

  • K. López de Ipiña
  • M. Graña
  • N. Ezeiza
  • M. Hernández
  • E. Zulueta
  • A. Ezeiza
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2905)

Abstract

This paper presents a new methodology, based on the classical decision trees, to get a suitable set of context dependent sublexical units for Basque Continuous Speech Recognition (CSR). The original method proposed by Bahl [1] was applied as the benchmark. Then two new features were added: a data massaging to emphasise the data and a fast and efficient Growing and Pruning algorithm for DT construction. In addition, the use of the new context dependent units to build word models was addressed. The benchmark Bahl approach gave recognition rates clearly outperforming those of context independent phone-like units. Finally the new methodology improves over the benchmark DT approach.

Keywords

Sublexical Units Decision Trees Growing and Pruning Algorithm 

References

  1. 1.
    Bahl, R.L., de Souza, V.P., Gopalakrishnan, P.S., Nahamoo, D., Picheny, M.A.: Decision Trees for Phonological Rules in Continuos Speech Recognition. In: Proc. IEEE ICASSP 1994, pp. 533–536 (1994)Google Scholar
  2. 2.
    Kuhn, R., Lazarides, A., Normandin, Y.: Improving Decision Trees for Phonetic Modelling. In: Proc. IEEE ICASSP 1995, pp. 552–555 (1995)Google Scholar
  3. 3.
    Odell, J.: The Use of Context in Large Vocabulary Speech Recognition. Ph. Thesis. Cambridge University (March 1995)Google Scholar
  4. 4.
    Gelfand, S.B., Ravishankar, C.S., Delp, E.J.: An Iterative Growing and Pruning Algorithm for Classification Tree Design. IEEE Trans. on PAMI 13(2), 163–174 (1991)Google Scholar
  5. 5.
    Breiman, L., Friedman, J.H., Olshen, R.A., Stone, C.J.: Classification and Regression Trees. Wadsworth & Brooks (1984)Google Scholar
  6. 6.
    Young, S.J., Odell, J., Woodland, P.: Tree-Based State Tying for High Accuracy Acoustic Modelling. In: ARPA Workshop on Human Language Technology, March 1994, pp. 286–291 (1994)Google Scholar
  7. 7.
    López de Ipiña, K., Varona, A., Torres, I., Rodríguez, L.J.: Decision Trees for Inter- Word Context Dependencies in Spanish Continuous Speech Recognition Tasks. In: EUROSPEECH 1999, Budapest (1999)Google Scholar
  8. 8.
    Feldman, J.A., Lakoff, A., Stolcke, A., Weber, S.H.: Miniature language Acquisition: a touch-stone for cognitive science. Technical Report, Tr-90-009. ICSI, Berkeley, California (April 1990)Google Scholar
  9. 9.
    Lopez de Ipiña, K., Torres, I., Oñederra, L., Hernandez, M., Peñagarikano, M., Ezeiza, N., Varona, A., Rodríguez, L.J.: First Selection of Lexical Units for Continuos Speech Recognition of Basque. In: Proceedings of ICSLP, Beijing, vol. II, pp. 531–535 (2000)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2003

Authors and Affiliations

  • K. López de Ipiña
    • 1
  • M. Graña
    • 2
  • N. Ezeiza
    • 3
  • M. Hernández
    • 2
  • E. Zulueta
    • 1
  • A. Ezeiza
    • 3
  1. 1.Sistemen Ingeniaritza eta Automatika Saila Gasteiz 
  2. 2.Konputazio Zientziak eta Adimen Artifiziala SailaDonostia
  3. 3.IXA groupUniversity of the Basque CountryDonostia

Personalised recommendations