Using Finite State Models for the Integration of Hierarchical LMs into ASR Systems

  • Raquel Justo
  • M. Inés Torres
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6718)


Through out this work we explore different methods to integrate a complex Language Model (a hierarchical Language Model based on classes of phrases) into an Automatic Speech Recognition (ASR) system. The integration is carried out by means of a composition of the different Stochastic Finite State Automata associated to the specific Language Model. This method is based on the same idea employed to integrate the different knowledge sources involved in the recognition process when a classical word-based Language Model is considered. The obtained results show that this integrated architecture provides better ASR system performance than a two-pass decoder where the complex LM is employed to reorder the N-best list.


stochastic finite state models speech recognition hierarchical language models 


  1. 1.
    Benedí, J., Lleida, E., Varona, A., Castro, M., Galiano, I., Justo, R., López, I., Miguel, A.: Design and acquisition of a telephone spontaneous speech dialogue corpus in Spanish: DIHANA. In: Proceedings of LREC 2006, Genoa, Italy (May 2006)Google Scholar
  2. 2.
    Benedí, J.M., Sánchez, J.A.: Estimation of stochastic context-free grammars and their use as language models. Computer Speech & Language 19(3), 249–274 (2005)CrossRefGoogle Scholar
  3. 3.
    Caseiro, D., Trancoso, I.: A specialized on-the-fly algorithm for lexicon and language model composition. IEEE Transactions on Audio, Speech & Language Processing 14(4), 1281–1291 (2006)CrossRefGoogle Scholar
  4. 4.
    García, P., Vidal, E.: Inference of k-testable languages in the strict sense and application to syntactic pattern recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence 12(9), 920–925 (1990)CrossRefGoogle Scholar
  5. 5.
    Jurafsky, D., Wooters, C., Segal, J., Stolcke, A., Fosler, E., Tajchman, G., Morgan, N.: Using a stochastic context-free grammar as a language model for speech recognition. In: Proceedings of ICASSP 1995, pp. 189–192. IEEE Computer Society Press, Detroit (1995)Google Scholar
  6. 6.
    Justo, R., Pérez, A., Torres, M.I.: Impact of the approaches involved on word-graph derivation from the asr system. In: Proceedings of the IbPRIA 2011, Las Palmas de Gran Canaria, Spain, June 8-10 (2011) (to be published in LNCS)Google Scholar
  7. 7.
    Justo, R., Torres, M.I.: Phrase classes in two-level language models for asr. Pattern Analysis & Applications 12(4), 427–437 (2009)MathSciNetCrossRefGoogle Scholar
  8. 8.
    Mohri, M., Riley, M.: A weight pushing algorithm for large vocabulary speech recognition. In: Proceedings of INTERSPEECH 2001, Aalborg, Denmark, September 2001, pp. 1603–1606 (2001)Google Scholar
  9. 9.
    Niesler, T., Whittaker, E., Woodland, P.: Comparison of part-of-speech and automatically derived category-based language models for speech recognition. In: ICASSP 1998, Seattle, pp. 177–180 (1998)Google Scholar
  10. 10.
    Pereira, F., Riley, M.D.: Speech recognition by composition of weighted finite automata. In: Finite-State Language Processing, pp. 431–453. MIT Press, Cambridge (1996)Google Scholar
  11. 11.
    Torres, M.I., Varona, A.: k-TSS language models in speech recognition systems. Computer Speech and Language 15(2), 127–149 (2001)CrossRefGoogle Scholar
  12. 12.
    Zitouni, I.: Backoff hierarchical class n-gram language models: effectiveness to model unseen events in speech recognition. Computer Speech and Language 21(1), 99–104 (2007)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  • Raquel Justo
    • 1
  • M. Inés Torres
    • 1
  1. 1.University of the Basque CountryLeioaSpain

Personalised recommendations