Speech Enhancement Via Correlation Coefficients

Part of the SpringerBriefs in Electrical and Computer Engineering book series (BRIEFSELECTRIC)


In the previous two chapters, we showed the importance of different kinds of correlation coefficients in the formulation and analysis of the best estimators for speech enhancement. In this chapter, we focus on the linear case and show how the most relevant noise reduction filters as well as new ones can be easily derived from the Pearson correlation coefficient. We work in the time domain but the extension of these ideas to the more convenient short-time Fourier transform domain is straightforward.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    J. Benesty, J. Chen, and Y. Huang, “On the importance of the Pearson correlation coefficient in noise reduction,” IEEE Trans. Audio, Speech, Language Process., vol. 16, pp. 757–765, May 2008.Google Scholar
  2. 2.
    J. Benesty, J. Chen, Y. Huang, and I. Cohen, Noise Reduction in Speech Processing. Berlin, Germany: Springer-Verlag, 2009.Google Scholar
  3. 3.
    J. Yu, J. Benesty, G. Huang, and J. Chen, “Examples of optimal noise reduction filters derived from the squared Pearson correlation coefficient,” in Proc. IEEE ICASSP, 2014, pp. 1571–1575.Google Scholar
  4. 4.
    J. Yu, J. Benesty, G. Huang, and J. Chen, “Optimal single-channel noise reduction filtering matrices from the Pearson correlation coefficient perspective,” in Proc. IEEE ICASSP, 2015, pp. 201–205.Google Scholar
  5. 5.
    J. N. Franklin, Matrix Theory. Englewood Cliffs, NJ: Prentice-Hall, 1968.Google Scholar
  6. 6.
    J. Benesty, J. Chen, Y. Huang, and T. Gaensler, “Time-domain noise reduction based on an orthogonal decomposition for desired signal extraction,” J. Acoust. Soc. Am., vol. 132, pp. 452–464, July 2012.Google Scholar
  7. 7.
    J. Benesty and Y. Huang, “A single-channel noise reduction MVDR filter,” in Proc. IEEE ICASSP, 2011, pp. 273–276.Google Scholar
  8. 8.
    A. Schasse and R. Martin, “Estimation of subband speech correlations for noise reduction via MVDR processing,” IEEE Trans. Audio, Speech, Language Process., vol. 22, pp. 1355-1365, Sept. 2014.Google Scholar
  9. 9.
    D. Fischer and T. Gerkmann, “Single-microphone speech enhancement using MVDR filtering and Wiener post-filtering,” in Proc. IEEE ICASSP, 2016, pp. 201–205.Google Scholar
  10. 10.
    D. Fischer, S. Doclo, E. A. P. Habets, and T. Gerkmann, “Combined single-microphone Wiener and MVDR filtering based on speech interframe correlations and speech presence probability,” in Proc. ITG Conf. Speech Communication, 2016, pp. 292–296.Google Scholar

Copyright information

© The Author(s) 2018

Authors and Affiliations

  1. 1.INRS EMT, Suite 6900University of QuebecMontrealCanada

Personalised recommendations