Speech Enhancement Based on the Response Features of Facilitated EI Neurons

  • André B. Cavalcante
  • Danilo P. Mandic
  • Tomasz M. Rutkowski
  • Allan Kardec Barros
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3889)


A real-time approach for the enhancement of speech at zero degree azimuth is proposed. This is achieved inspired by the response features of the “Facilitated EI neurons”. This way, frequency segregation through a bandpass filter bank is followed by “supression analysis” which inhibits sources that are not at “facilitated” positions. Unlike with the existing approaches for the solution of cocktail party problem, where the performance under low SNR (signal-to-noise ratio) reverberation conditions is severely limited, the proposed approach has the capability to circumvent these problems. This is quantified through both objective and subjective performance measures and supported by real world simulation examples.


Inferior Colliculus Blind Source Separation Speech Enhancement Reverberation Time Human Auditory System 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Cherry, E.C.: Some experiments on the recognition of speech, with one and two ears. Journal of the Acoustic Society of America 25, 975–979 (1953)CrossRefGoogle Scholar
  2. 2.
    Hyvarinen, A., Oja, E.: A fast fixed-point algorithm for independent component analysis. Neural Computation 9, 1483–1492 (1997)CrossRefGoogle Scholar
  3. 3.
    de Cheveigne, A.: The auditory system as a separation machine. In: Proceedings of International Symposium of Hearing (2000)Google Scholar
  4. 4.
    Barros, A.K., Ohnishi, N.: Single channel speech enhancement by efficient coding. Signal Processing 85, 1805–1812 (2005)MATHCrossRefGoogle Scholar
  5. 5.
    Lewicki, M.: Efficient coding of natural sounds. Nature Neuroscience 5(4), 356–363 (2002)CrossRefGoogle Scholar
  6. 6.
    Roman, N., Wang, D., Brown, G.: Speech segregation based on sound localization. Journal of Acoustical Society of America 114(1), 2236–2252 (2003)CrossRefGoogle Scholar
  7. 7.
    Virgag, N.: Single channel speech enhancement based on masking properties of the human auditory system. IEEE Transactions on Signal Processing 7, 126–137 (1999)Google Scholar
  8. 8.
    Barros, A.K., Rutkowski, T.M., 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) (2002)Google Scholar
  9. 9.
    Pollak, G., Burger, R., Park, T., Klug, A., Bauer, E.: Roles of inhibition for transforming binaural properties in the brainstem auditory system. Hearing Research 168(1–2), 60–78 (2002)CrossRefGoogle Scholar
  10. 10.
    Pollak, G., Burger, R., Klug, A.: Dissecting the circuitry of the auditory system. Trends in Neuroscience 26, 33–39 (2004)CrossRefGoogle Scholar
  11. 11.
    Grantham, G.: Discrimination of dynamic interaural intensity differences. Journal of Acoustical Society of America 76(1), 71–76 (1984)CrossRefGoogle Scholar
  12. 12.
    Allen, J., Berkley, D.: Image method for efficiently simulating small-room acoustics. Journal of Acoustical Society of America 65, 943–950 (1979)CrossRefGoogle Scholar
  13. 13.
    Hansen, J., Pellom, B.L.: An effective qualiy evaluation protocol for speech enhancement algorithms. In: Proceedings of ICSLP 1998, vol. 7, pp. 2819–2822 (1998)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • André B. Cavalcante
    • 1
  • Danilo P. Mandic
    • 2
  • Tomasz M. Rutkowski
    • 3
  • Allan Kardec Barros
    • 1
  1. 1.Laboratory for Biological Information ProcessingUniversidade Federal do MaranhāoBrazil
  2. 2.Department of Electrical and Electronic EngineeringImperial College LondonUnited Kingdom
  3. 3.Laboratory for Advanced Brain Signal ProcessingBrain Science Institute RikenJapan

Personalised recommendations