Speech Segregation Using Constrained ICA

  • Qiu-Hua Lin
  • Yong-Rui Zheng
  • Fuliang Yin
  • Hua-Lou Liang
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3173)


In natural environment, speech often occurs concurrently with acoustic interference. How to effectively extract speech remains a great challenge. This paper describes a novel constrained Independent Component Analysis (ICA) approach, the ICA with reference (ICA-R), to speech segregation. Different from the traditional ICA which recovers simultaneously all the source signals, the ICA-R extracts only some desired source signals from the mixtures of source signals by incorporating some a priori information into the separation process. We show how the ICA-R can be applied to separate a target speech signal from interfering sounds by exploiting a proper reference signal, which is based on the different characteristic between speech signal and its environmental noises, i.e., the speech signal has pitch and its harmonic frequencies whereas the noises usually do not. Results of computer experiments demonstrate the efficiency of the proposed method.


Independent Component Analysis Speech Signal Reference Signal Independent Component Analysis Power Density Spectrum 
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.
    Comon, P.: Independent component analysis, a new concept? Signal Processing 36(3), 287–314 (1994)MATHCrossRefGoogle Scholar
  2. 2.
    Cardoso, J.F.: Blind signal separation: statistical principles. Proc. of the IEEE 86(10), 2009–2025 (1998)CrossRefGoogle Scholar
  3. 3.
    Amari, S., Cichocki, A.: Adaptive blind signal processing - Neural network approaches. Proc. of the IEEE 86(10), 2026–2048 (1998)CrossRefGoogle Scholar
  4. 4.
    Hyvärinen, A., Oja, E.: Independent component analysis: algorithms and applications. Neural Networks 13(4-5), 411–430 (2000)CrossRefGoogle Scholar
  5. 5.
    Barros, A.K., Rutkowski, T., Itakura, F., Ohnishi, N.: Estimation of speech embedded in a reverberant and noisy environment by independent component analysis and wavelets. IEEE Transactions on Neural Networks 13(4), 888–893 (2002)CrossRefGoogle Scholar
  6. 6.
    Visser, E., Lee, T.W.: Speech enhancement using blind source separation and two-channel energy based speaker detection. In: Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2003), vol. 1, pp. 884–887 (2003)Google Scholar
  7. 7.
    Lee, J.H., Jung, H.Y., Lee, T.W., Lee, S.Y.: Speech enhancement with MAP estimation and ICA-based speech features. Electronics Letters 36(17), 1506–1507 (2000)CrossRefGoogle Scholar
  8. 8.
    Lu, W., Rajapakse, J.C.: ICA with reference. In: Proc. Third Int. Conf. on ICA and Blind Source Separation (ICA 2001), pp. 120–125 (2001)Google Scholar
  9. 9.
    Hyvärinen, A.: New approximations of differential entropy for independent component analysis and projection pursuit. Advances in Neural Information Processing Systems 10, 273–279 (1998)Google Scholar
  10. 10.
    Cooke, M.P.: Modeling Auditory Processing and Organization. Cambridge Univ. Press, Cambridge (1993)Google Scholar
  11. 11.
    Tabrikian, J., Dubnov, S., Dickalov, Y.: Maximum a-posteriori probability pitch tracking in noisy environments using harmonic model. IEEE Transactions on Speech and Audio Processing 12(1), 76–87 (2004)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2004

Authors and Affiliations

  • Qiu-Hua Lin
    • 1
  • Yong-Rui Zheng
    • 1
  • Fuliang Yin
    • 1
  • Hua-Lou Liang
    • 2
  1. 1.School of Electronic and Information EngineeringDalian University of TechnologyDalianChina
  2. 2.School of Health Information SciencesThe University of Texas at HoustonHoustonUSA

Personalised recommendations