Integrating Phonological Knowledge in ASR Systems for Spanish Language

  • Javier Mikel Olaso
  • María Inés Torres
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6419)


In this paper we undertake the use of phonological features applied to speech recognition in Spanish language. We investigate two different ways to integrate these phonological features into an HMM based speech recognition system. We also propose a method to integrate these features using an architecture that uses independent feature streams. In the experimental results we find that higher recognition accuracies and less computational cost can be obtained.


speech recognition acoustic modeling phonological features 


  1. 1.
    Rose, R., Momayyez, P.: Integration of multiple feature sets for reducing ambiguity in ASR. In: ICASSP 2007, vol. 4, pp. 325–328 (2007)Google Scholar
  2. 2.
    Rose, R., et al.: An investigation of the potential role of speech production models in automatic speech recognition. In: Proceedings ICSLP 1994, pp. 575–578 (1994)Google Scholar
  3. 3.
    Koreman, J., Andreeva, B.: Can we use the linguistic information in the signal? Phonus (Institute of Phonetics, University of the Saarland) 5, 47–58 (2000)Google Scholar
  4. 4.
    Livescu, K., et al.: Articulatory Feature-based Methods for Acoustic and Audio-Visual Speech Recognition: 2006 JHU Summer Workshop Final Report, Technical Report, Center for Language and Speech Processing, Johns Hopkins University (2007)Google Scholar
  5. 5.
    Strom, N.: Phoneme probability estimation with dynamic sparsely connected artificial neural networks. The Free Speech Journal 1(#5) (1997)Google Scholar
  6. 6.
    Blackburn, C.S., Young, S.J.: Pseudo-Articulatory speech synthesis for recognition using automatic feature extraction from X-Ray data. In: Proceedings ICSLP 1996, pp. 969–972 (1996)Google Scholar
  7. 7.
    Saenko, K., et al.: Articulatory features for robust visual speech recognition. In: ICMI 2004 (2004)Google Scholar
  8. 8.
    King, S., Taylor, P.: Detection of phonological features in continuous speech using neural networks. Computer Speech & Language, 333–353 (2000)Google Scholar
  9. 9.
    Frankel, J., et al.: Articulatory feature recognition using dynamic Bayesian networks. Computer Speech and Language Archive 21(4), 620–640 (2007)CrossRefGoogle Scholar
  10. 10.
    Parya, M., et al.: Exploiting complementary aspects of phonological features in automatic speech recognition. In: IEEE Workshop on Automatic Speech Recognition & Understanding, pp. 47–52 (2007)Google Scholar
  11. 11.
    Stouten, F., Martens, J.P.: On the use of phonological features for pronunciation scoring. In: Proceedings ICASSP, pp. 229–232 (2006)Google Scholar
  12. 12.
    BenZeghiba, M., et al.: Automatic speech recognition and intrinsic speech variation. In: 31st International Conference on Acoustics, Speech, and Signal Processing ICASSP 2006, May 14-19 (2006)Google Scholar
  13. 13.
    Abu-Amer, T., Carson-Berndsen, J.: HARTFEX: A multi-dimentional system of HMM based recognisers for articulatory features extraction. In: Proceedings NOLISP 2003, paper009 (2003)Google Scholar
  14. 14.
    Casacuberta, F., et al.: Desarrollo de corpus para investigación en tecnologías del habla (Albayzin). Procesamiento del Lenguage Natural 12, 35–42 (1992)Google Scholar
  15. 15.
    Olaso, J.M., Torres, M.I.: Speech production models for ASR in Spanish language. Paper Submmited to FALA 2010 (2010)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2010

Authors and Affiliations

  • Javier Mikel Olaso
    • 1
  • María Inés Torres
    • 1
  1. 1.Universidad del País VascoSpain

Personalised recommendations