Continuous Speech Recognition Based on ICA and Geometrical Learning

  • Hao Feng
  • Wenming Cao
  • Shoujue Wang
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3930)


We investigate the use of independent component analysis (ICA) for speech feature extraction in digits speech recognition systems. We observe that this may be true for recognition tasks based on Geometrical Learning with little training data. In contrast to image processing, phase information is not essential for digits speech recognition. We therefore propose a new scheme that shows how the phase sensitivity can be removed by using an analytical description of the ICA-adapted basis functions. Furthermore, since the basis functions are not shift invariant, we extend the method to include a frequency-based ICA stage that removes redundant time shift information. The digits speech recognition results show promising accuracy. Experiments show that the method based on ICA and Geometrical Learning outperforms HMM in a different number of training samples.


Discrete Cosine Transform Recognition Rate Independent Component Analysis Speech Recognition Speech Signal 
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.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Amari, S.: Neural learning in structured parameter spaces—natural Riemannian gradient. In: Advances in Neural Information Processing System, vol. 9, pp. 127–133. MIT Press, Cambridge (1997)Google Scholar
  2. 2.
    Bell, A., Sejnowski, T.: An information-maximization approach to blind separation and blind deconvolution. Neural Comput. 7, 1129–1159 (1995)CrossRefGoogle Scholar
  3. 3.
    Bell, A.J., Sejnowski, T.J.: Learning the higher-order structure of a natural sound. Network Comput. Neural Syst. 7, 261–266 (1996)MATHCrossRefGoogle Scholar
  4. 4.
    Bell, A.J., Sejnowski, T.J.: The ‘independent components’ of natural scenes are edge filters. Vision Res. 37(23), 3327–3338 (1997)CrossRefGoogle Scholar
  5. 5.
    Box, G.E.P., Tiao, G.C.: Bayesian Inference in Statistical Analysis. Wiley, New York (1992)MATHGoogle Scholar
  6. 6.
    Lee, J.H., Jung, H.Y., Lee, T.W., Lee, S.Y.: Speech feature extraction using independent component analysis. In: Proceedings of the International Conference Acoustics, Speech, Signal Processing, Istanbul, Turkey, June 2000, pp. 1631–1634 (2000)Google Scholar
  7. 7.
    Lee, J.-H., Lee, T.-W., Jung, H.-Y., Lee, S.-Y.: On the efficient speech feature extraction based on independent component analysis. Neural Process. Lett. 15(3), 235–245 (2002)MATHCrossRefGoogle Scholar
  8. 8.
    ShouJue, W.: A new development on ANN in China - Biomimetic pattern recognition and multi weight vector neurons. LNCS (LNAI), vol. 2639, pp. 35–43. Springer, Heidelberg (2003)Google Scholar
  9. 9.
    Shoujue, W., et al.: Multi Camera Human Face Personal Identification System Based on Biomimetic pattern recognition. Acta Electronica Sinica 31(1), 1–3 (2003)Google Scholar
  10. 10.
    Shoujue, W., et al.: Discussion on the basic mathematical models of Neurons in General purpose Neurocomputer. Acta Electronica Sinica 29(5), 577–580 (2001)Google Scholar
  11. 11.
    Wang, X., Wang, S.: The Application of Feedforward Neural Networks in VLSI Fabrication Process Optimization. International Journal of Computational Intelligence and Applications 1(1), 83–90 (2001)CrossRefGoogle Scholar
  12. 12.
    Cao, W., Hao, F., Wang, S.: The application of DBF neural networks for object recognition. Inf. Sci. 160(1-4), 153–160 (2004)CrossRefGoogle Scholar
  13. 13.
    Hyvärinen, A., Karhunen, J., Oja, E.: Independent Component Analysis. Wiley, New York (2001)CrossRefGoogle Scholar
  14. 14.
    Cao, W.M.: Similarity index for clustering DNA microarray data based on multi-weighted neuron. In: Ślęzak, D., Yao, J., Peters, J.F., Ziarko, W.P., Hu, X. (eds.) RSFDGrC 2005. LNCS (LNAI), vol. 3642, pp. 402–408. Springer, Heidelberg (2005)CrossRefGoogle Scholar
  15. 15.
    Cao, W.M., Hu, J.H., Xiao, G., et al.: Application of multi-weighted neuron for iris recognition. In: Wang, J., Liao, X.-F., Yi, Z. (eds.) ISNN 2005. LNCS, vol. 3497, pp. 87–92. Springer, Heidelberg (2005)CrossRefGoogle Scholar
  16. 16.
    Cao, W.M.: The application of Direction basis function neural networks to the prediction of chaotic time series. Chinese Journal of Electronics 13(3), 395–398 (2004)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Hao Feng
    • 3
  • Wenming Cao
    • 1
    • 2
  • Shoujue Wang
    • 2
  1. 1.Institute of Intelligent Information System, Information CollegeZhejiang University of TechnologyHangzhouChina
  2. 2.Institute of SemiconductorsChinese Academy of ScienceBeijingChina
  3. 3.Jiaxing UniversityChina

Personalised recommendations