The Improved Method for Solving Permutation Problem in Frequency Domain Blind Source Separation of Speech Signals

Abstract:

The signals of convolutive mixture in time-domain can be transformed to instantaneous mixtures in frequency-domain and complex-valued independent component analysis (CICA) can separate efficiently the signals of instantaneous mixture in each frequency bin. However, since CICA is calculated in each frequency bin independently, the permutation ambiguity becomes a serious problem. The permutation ambiguity of CICA in each frequency bin should be aligned so that a separated signal in the time-domain contains frequency components of the same source signal. The paper presents a novel and efficient approach for solving the permutation problem in frequency domain blind source separation of speech signals. The new algorithm models the frequency-domain separated signals by means of Teager energy correlation between neighboring bins for the detection of correct permutations. Experimental results show that the proposed algorithm can efficiently solve the permutation ambiguity problem in each frequency bin.

• Periodical: Advanced Materials Research (Volumes 433-440)
• Pages: 7029-7034
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.433-440.7029
• Online since: January 2012
• Authors: De Xiang Zhang, Xiao Pei Wu, Zhao Lv, Xiao Jing Guo
• Keywords: Blind Source Separation, Independent Component Analysis (ICA), Permutation, Teager Energy Correlation

References

[1] P. Comon, Independent component analysis, A new concept?, Signal Process., vol. 36, no. 3, p.287–314, (1994).

DOI: 10.1016/0165-1684(94)90029-9

[2] E. Bingham and A. Hyvarinen, A fast fixed-point algorithm for independent component analysis of complex-valued signals, Int. J. of Neural Systems, vol 10, no 1, pp.1-8, Jan. (2000).

DOI: 10.1142/s0129065700000028

[3] S. Cruces-Alvarez, A. Cichocki, and S. Amari, From blind signal extraction to blind instantaneous signal separation: Criteria, algorithms, and stability, IEEE Trans. Neural Netw., vol. 15, no. 4, p.859–873, Jul. (2004).

DOI: 10.1109/tnn.2004.828764

[4] N. Murata, S. Ikeda and A. Ziehe, An approach to blind source separation based on temporal structure of speech signals, Neurocomput., vol. 41, no. 1-4, pp.1-24, Oct. (2001).

DOI: 10.1016/s0925-2312(00)00345-3

[5] H. Sawada, R. Mukai, S. Araki, S. Makino, A robust and precise method for solving the permutation problem of frequency-domain blind source separation, IEEE Transactions on Speech and Audio Processing, Vol. 12, Issue 5, pp.530-538, Sept. (2004).

DOI: 10.1109/tsa.2004.832994

[6] M. Z. Ikram and D. R. Morgan, Permutation inconsistency in blind speech separation: Investigation and solutions, IEEE Trans. Speech Audio Process., vol. 13, no. 1, p.1–13, Jan. (2005).

DOI: 10.1109/tsa.2004.834441

[7] S. C. Douglas, H. Sawada, and S. Makino, Natural gradient multichannel blind deconvolution and speech separation using causal FIR filters, IEEE Trans. Speech Audio Process., vol. 13, no. 1, p.92–104, Jan. (2005).

DOI: 10.1109/tsa.2004.838538

[8] J. F. Kaiser, On a simple algorithm to calculate the 'energy' of a signal, in Proc. IEEE International Conference on Acoust., Speech, Signal Processing, Albuquerque, New Mexico, USA, Apr. 1990, p.381–384.

[9] R. Mazur, A. Mertins, and M. E. Davies. Solving the permutation problem in convolutive blind source separation, in Independent Component Analysis and Signal Separation. New York: Springer, 2007, vol. 4666, p.512–519.

DOI: 10.1007/978-3-540-74494-8_64

[10] T. Mei, J. Xi, F. Yin, A. Mertins, and J. F. Chicharo, Blind source separation based on time-domain optimizations of a frequency-domain independence criterion, IEEE Trans. Audio, Speech, Lang. Process., vol. 14, no. 6, p.2075–2085, Nov. (2006).

DOI: 10.1109/tasl.2006.872623