Blind Source Separation for Convoluted Signals Based on Properties of Acoustic Transfer Function in Real Environments

Authors: Takaaki Ishibashi

Abstract:

Frequency domain independent component analysis has a scaling indeterminacy and a permutation problem. The scaling indeterminacy can be solved by use of a decomposed spectrum. For the permutation problem, we have proposed the rules in terms of gain ratio and phase difference derived from the decomposed spectra and the source-s coarse directions. The present paper experimentally clarifies that the gain ratio and the phase difference work effectively in a real environment but their performance depends on frequency bands, a microphone-space and a source-microphone distance. From these facts it is seen that it is difficult to attain a perfect solution for the permutation problem in a real environment only by either the gain ratio or the phase difference. For the perfect solution, this paper gives a solution to the problems in a real environment. The proposed method is simple, the amount of calculation is small. And the method has high correction performance without depending on the frequency bands and distances from source signals to microphones. Furthermore, it can be applied under the real environment. From several experiments in a real room, it clarifies that the proposed method has been verified.

Keywords: blind source separation, frequency domain independent component analysys, permutation correction, scale adjustment, target extraction.

Digital Object Identifier (DOI): doi.org/10.5281/zenodo.1081649

References:

[1] A. J. Bell and T. J. Sejnowski: An information maximization approach to blind separation and blind deconvolution, Neural Computation, Vol. 7, No. 6, pp. 1129-1159, 1995.
[2] A. Cichocki and S. Amari: Adaptive blind signal and image processing, Learning algorithm and applications, John Wiley & Sons, Ltd, 2002.
[3] T. W. Lee, M. Girolami and T. J. Sejnowski: Independent component analysis using an extended informax algorithm for mixed subgaussian and supergaussian sources, Neural Computation, Vol. 11, No. 2, pp. 417-441, 1999.
[4] A. Hyv¨arinen, J. Karhunen and E. Oja, "Independent component analysis," John Wiley & Sons, Ltd, 2001.
[5] N. Murata, S. Ikeda and A. Ziehe: An approach to blind source separation based on temporal structure of speech signals, Neurocomputing, Vol. 41, Issue 1-4, pp. 1-24, 2001.
[6] H. Gotanda, K. Nobu, T. Koya, K. Kaneda, T. Ishibashi and N. Haratani: Permutation correction and speech extraction based on split spectrum through FastICA, 4th International Symposium on Independent Component Analysis and Blind Signal Separation (ICA2003), pp. 379- 384, 2003.
[7] H. Saruwatari, S. Kurita, K. Takeda, F. Itakura, T. Nishikawa and K. Shikano: Blind source separation combining independent component analysis and beamforming, EURASIP Journal on Applied Signal Processing, Vol.2003, No.11, pp.1135-1146, 2003.
[8] Acoustical Society of Japan: ASJ continuous speech corpus japanese newspaper article sentences, JNAS Vols.1-16, 1997.
[9] NTT Advanced Technology Corporation: Ambient noise database for telephonometry 1996, 1996.
[10] T. Koya, N. Iwasaki, T. Ishibashi, G. Hirano, H. Shiratsuchi and Hiromu Gotanda, "SN ratio estimation and speech segment detection of extracted signals through Independent Component Analysis," Journal of Advanced Computational Intelligence and Intelligent Informatics, Vol. 14, No. 4, pp.364-374, 2010.