The Journal of the Acoustical Society of Korea

The Acoustical Society of Korea (한국음향학회)
권호 표지

A New Formulation of Multichannel Blind Deconvolution: Its Properties and Modifications for Speech Separation

  • Nam, Seung-Hyon (Department of Electronic Engineering, Paichai University) ;
  • Jee, In-Nho (Department of Computer and Information Communication, Hongik University)
  • Published : 2006.12.30
Download PDF
⟨ Previous Next ⟩

Abstract

A new normalized MBD algorithm is presented for nonstationary convolutive mixtures and its properties/modifications are discussed in details. The proposed algorithm normalizes the signal spectrum in the frequency domain to provide faster stable convergence and improved separation without whitening effect. Modifications such as nonholonomic constraints and off-diagonal learning to the proposed algorithm are also discussed. Simulation results using a real-world recording confirm superior performanceof the proposed algorithm and its usefulness in real world applications.

Keywords

References

  1. S.-I. Amari, S. C. Douglas, A. Cichocki, and H. H. Yang, 'Novel on-line adaptive learning algorithms for blind deconvolution using the natural gradient approach,' Proc. IEEE 11th IFAC Symposium on System Identification, SYSID-97, Kitakyushu, Japan, 1057-1062, 1997,
  2. T.-W. Lee, A. J. Bell, and R. Lambert, 'Blind separation of delayed and convolved sources,' Advances in Neural Information Processing Systems, MIT Press, 9 758-764, 1997
  3. S. H. Nam and S. Beack, 'A frequency-domain normalized multichannel blind deconvolution algorithm for acoustical signals,' Proc. lnt. Workshop on Independent Component Analysis and Signal Separation (ICA'04), Granada, Spain, 522-529, 2004,
  4. S. Beack, S. H. Nam, and M. Hahn, 'A new speech enhancement algorithm for car environment noise cancellation with MBD and Kalman filtering,' IEICE Trans, Fundamentals, vol. E88-A, 3 685-689, March 2005, https://doi.org/10.1093/ietfec/e88-a.3.685
  5. J.-F. Cardoso, 'Blind signal separation: statistical principles,' Proceedings of the IEEE, 9(10) 2009-2025, 1998, https://doi.org/10.1109/5.720250
  6. S.-I. Amari, 'Natural gradient works efficiently in learning,' Neural Computation, 10(2) 251-276, 1998 https://doi.org/10.1162/089976698300017746
  7. L. Zhang, A. Cichocki, and S. Amari, 'Geometrical structures of FIR manifold and their application to multichannel blind deconvolution,' Proc. IEEE Workshop on Neural Networks for Signal Processing (NNSP'99), Madison, Wisconsin, 303-312, 1999,
  8. M. Joho and P. Schniter, 'Frequency domain realization of multichannel blind deconvolution algorithm based on the natural gradient,' Proc. lnt. Workshop on Independent Component Analysis and Signal Separation (ICA'03), 2003
  9. C. Fancourt and L. Parra, 'The generalized sidelobe decorrelator,' IEEE Workshop on Applications of Signal Processing to Audio and Acoustics, 167-170, Oct. 2001
  10. H. Buchner, R. Aichner, and W. Kellermann, 'A generalization of a class of blind source separation algorithms for convolutional mixtures,' Proc. lnt. Workshop on Independent Component Analysis and Signal Separation (ICA'03), 2003
  11. S.-I. Amari, T.-P. Chen, and A.Cichocki, 'Nonholonomic orthogonal learning Algorithms for blind source separation,' Neural Computation, 12(6) 1463-1484, 2000 https://doi.org/10.1162/089976600300015466
  12. C. Fancourt and L. Parra, 'The coherence function in blind source separation of convolutive mixtures of non-stationary signals,' Proc IEEE Workshop on Neural Networks for Signal Processing, 303-312, 2001
  13. H. Buchner, R. Aichner, and W. Kellermann, 'Blind source separation for convolutional mixtures: A unified treatment,' in Audio Signal Processing: For next-generation multimedia communication systems, Y. Huang and J, Benesty (Eds), Kluwer Academic Pub., 2004