Skip to main content

Speech Source Separation in Convolutive Environments Using Space-Time-Frequency Analysis

Abstract

We propose a new method for speech source separation that is based on directionally-disjoint estimation of the transfer functions between microphones and sources at different frequencies and at multiple times. The spatial transfer functions are estimated from eigenvectors of the microphones' correlation matrix. Smoothing and association of transfer function parameters across different frequencies are performed by simultaneous extended Kalman filtering of the amplitude and phase estimates. This approach allows transfer function estimation even if the number of sources is greater than the number of microphones, and it can operate for both wideband and narrowband sources. The performance of the proposed method was studied via simulations and the results show good performance.

References

  1. 1.

    Torkkola K: Blind separation for audio signals—are we there yet? Proceedings of 1st International Workshop on Independent Component Analysis and Blind Signal Separation (ICA~'99), January 1999, Aussois, France 239–244.

    Google Scholar 

  2. 2.

    Parra L, Spence C: Convolutive blind separation of non-stationary sources. IEEE Transactions on Speech and Audio Processing 2000, 8(3):320–327. 10.1109/89.841214

    Article  Google Scholar 

  3. 3.

    Jourjine A, Rickard S, Yilmaz O: Blind separation of disjoint orthogonal signals: demixing N sources from 2 mixtures. Proceedings of IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP '00), June 2000, Istanbul, Turkey 5: 2985–2988.

    Google Scholar 

  4. 4.

    Roman N, Wang DL, Brown GJ: Speech segregation based on sound localization. The Journal of the Acoustical Society of America 2003, 114(4):2236–2252. 10.1121/1.1610463

    Article  Google Scholar 

  5. 5.

    Fevotte C, Doncarli C: Two contributions to blind source separation using time-frequency distributions. IEEE Signal Processing Letters 2004, 11(3):386–389. 10.1109/LSP.2003.819343

    Article  Google Scholar 

  6. 6.

    Deville Y: Temporal and time-frequency correlation-based blind source separation methods. Proceedings of 4th International Workshop on Independent Component Analysis and Blind Signal Separation (ICA '03), April 2003, Nara, Japan 1059–1064.

    Google Scholar 

  7. 7.

    Ikram MZ, Morgan DR: Permutation inconsistency in blind speech separation: investigation and solutions. IEEE Transactions on Speech and Audio Processing 2005, 13(1):1–13.

    Article  Google Scholar 

  8. 8.

    Yilmaz O, Rickard S: Blind separation of speech mixtures via time-frequency masking. IEEE Transactions on Signal Processing 2004, 52(7):1830–1847. 10.1109/TSP.2004.828896

    MathSciNet  Article  Google Scholar 

  9. 9.

    Steinhardt A: Adaptive multisensor detection and estimation. In Adaptive Radar Detection and Estimation. Edited by: Haykin S, Steinhardt A. John Wiley & Sons, New York, NY, USA; 1992:91–160.

    Google Scholar 

  10. 10.

    Schobben DWE, Torkkola K, Smaragdis P: Evaluation of blind signal separation methods. Proceedings of 1st International Workshop on Independent Component Analysis and Blind Signal Separation (ICA '99), January 1999, Aussois, France

    Google Scholar 

Download references

Author information

Affiliations

Authors

Corresponding author

Correspondence to Shlomo Dubnov.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Dubnov, S., Tabrikian, J. & Arnon-Targan, M. Speech Source Separation in Convolutive Environments Using Space-Time-Frequency Analysis. EURASIP J. Adv. Signal Process. 2006, 038412 (2006). https://doi.org/10.1155/ASP/2006/38412

Download citation

Keywords

  • Information Technology
  • Transfer Function
  • Correlation Matrix
  • Multiple Time
  • Quantum Information