Blind Source Separation based on the Estimation for the Number of the Blind Sources under a Dynamic Acoustic Environment
Independent component analysis can estimate unknown
source signals from their mixtures under the assumption that the
source signals are statistically independent. However, in a real environment,
the separation performance is often deteriorated because
the number of the source signals is different from that of the sensors.
In this paper, we propose an estimation method for the number of
the sources based on the joint distribution of the observed signals
under two-sensor configuration. From several simulation results, it
is found that the number of the sources is coincident to that of
peaks in the histogram of the distribution. The proposed method can
estimate the number of the sources even if it is larger than that of
the observed signals. The proposed methods have been verified by
several experiments.
[1] A. Cichocki and S. Amari, Adaptive blind signal and image processing,
Learning algorithm and applications John Wiley & Sons, Ltd, 2002.
[2] A. Hyv¨arinen, J. Karhunen and E. Oja, Independent component analysis
John Wiley & Sons, Ltd, 2001.
[3] S. Amari, Natural gradient works efficiently in learning Neural
Computation, Vol. 10, pp. 251-276, 1998.
[4] 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.
[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] S. Ikeda and N. Murata, A method of ICA in time-frequency domain
International Workshop on Independent Component Analysis and Blind
Signal Separation (ICA-99), pp. 365-371, 1999.
[7] H. Saruwatari, H. Yamajo, T. Takatani, T. Nishikawa, and K. Shikano,
Blind separation and deconvolution for convolutive mixture of speech
combining SIMO-model-based ICA and multichannel inverse filtering
IEICE Trans. Fundamentals, Vol. E88-A, No. 9, pp. 2387-2400, 2005.
[8] T. Koya, T. Ishibashi, H. Shiratsuchi and H. Gotanda, Blind source
deconvolution based on frequency domain convolution model under
highly reverberant environments Transactions of the Institute of
Systems, Control and Information Engineers, Vol. 22, No. 8, pp. 287-
294, 2009.
[9] 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.
[10] T. Ishibashi, K. Inoue, H. Gotanda and K. Kumamaru, Frequency
domain independent component analysis without permutation and scale
indeterminacy Proceedings of the 41st ISCIE International Symposium
on Stochastic Systems Theory and Its Applications, pp. 190-195, 2009.
[11] H. Sawada, R. Mukai, S. Araki and S. Makino, Estimating the number
of sources using independent component analysis Acoustical Science
and Technology, the Acoustical Society of Japan, Vol. 26, No. 5, pp.
450-452, 2005.
[12] Acoustical Society of Japan, ASJ continuous speech corpus Japanese
newspaper article sentences JNAS Vols.1-16, 1997.
[13] NTT Advanced Technology Corporation, Ambient noise database for
telephonometry 1996 1996
[1] A. Cichocki and S. Amari, Adaptive blind signal and image processing,
Learning algorithm and applications John Wiley & Sons, Ltd, 2002.
[2] A. Hyv¨arinen, J. Karhunen and E. Oja, Independent component analysis
John Wiley & Sons, Ltd, 2001.
[3] S. Amari, Natural gradient works efficiently in learning Neural
Computation, Vol. 10, pp. 251-276, 1998.
[4] 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.
[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] S. Ikeda and N. Murata, A method of ICA in time-frequency domain
International Workshop on Independent Component Analysis and Blind
Signal Separation (ICA-99), pp. 365-371, 1999.
[7] H. Saruwatari, H. Yamajo, T. Takatani, T. Nishikawa, and K. Shikano,
Blind separation and deconvolution for convolutive mixture of speech
combining SIMO-model-based ICA and multichannel inverse filtering
IEICE Trans. Fundamentals, Vol. E88-A, No. 9, pp. 2387-2400, 2005.
[8] T. Koya, T. Ishibashi, H. Shiratsuchi and H. Gotanda, Blind source
deconvolution based on frequency domain convolution model under
highly reverberant environments Transactions of the Institute of
Systems, Control and Information Engineers, Vol. 22, No. 8, pp. 287-
294, 2009.
[9] 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.
[10] T. Ishibashi, K. Inoue, H. Gotanda and K. Kumamaru, Frequency
domain independent component analysis without permutation and scale
indeterminacy Proceedings of the 41st ISCIE International Symposium
on Stochastic Systems Theory and Its Applications, pp. 190-195, 2009.
[11] H. Sawada, R. Mukai, S. Araki and S. Makino, Estimating the number
of sources using independent component analysis Acoustical Science
and Technology, the Acoustical Society of Japan, Vol. 26, No. 5, pp.
450-452, 2005.
[12] Acoustical Society of Japan, ASJ continuous speech corpus Japanese
newspaper article sentences JNAS Vols.1-16, 1997.
[13] NTT Advanced Technology Corporation, Ambient noise database for
telephonometry 1996 1996
@article{"International Journal of Electrical, Electronic and Communication Sciences:62255", author = "Takaaki Ishibashi", title = "Blind Source Separation based on the Estimation for the Number of the Blind Sources under a Dynamic Acoustic Environment", abstract = "Independent component analysis can estimate unknown
source signals from their mixtures under the assumption that the
source signals are statistically independent. However, in a real environment,
the separation performance is often deteriorated because
the number of the source signals is different from that of the sensors.
In this paper, we propose an estimation method for the number of
the sources based on the joint distribution of the observed signals
under two-sensor configuration. From several simulation results, it
is found that the number of the sources is coincident to that of
peaks in the histogram of the distribution. The proposed method can
estimate the number of the sources even if it is larger than that of
the observed signals. The proposed methods have been verified by
several experiments.", keywords = "blind source separation, independent component analysys, estimation for the number of the blind sources, voice activity detection, target extraction.", volume = "4", number = "9", pages = "1408-6", }