Abstract
We propose a new algorithm with a stable learning and low-distortion based on overdetermined blind separation for the convolutive mixture of the speech. To improve the separation performance, we have proposed multistage ICA, in which frequency-domain ICA and time-domain ICA (TDICA) are cascaded. For temporally correlated signals, we must use TDICA with a nonholonomic constraint to avoid the decorrelation effect. However, the stability cannot be guaranteed in the nonholonomic case. Also, in the holonomic case, the sound quality of the separated signal is distorted by the decorrelation effect. To solve the problem of the stability, we perform TDICA with the holonomic constraint. To avoid the distortions, we estimate the distortion components by TDICA with the holonomic constraint and we compensate the sound qualities by using the estimated components. The stability of the proposed algorithm can be guaranteed by the holonomic constraint, and the proposed compensation work prevents the distortion. The experiments in a reverberant room reveal that the algorithm results in higher stability and higher separation performance.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Comon, P.: Independent component analysis, a new concept? Signal Processing 36, 287–314 (1998)
Amari, S., Douglas, S., Cichocki, A., Yang, H.: Multichannel blind deconvolution and equalization using the natural gradient In: Proc. SPAWC 1997, pp. 101–104 (1997)
Murata, N., Ikeda, S.: An on-line algorithm for blind source separation on speech signals In: Proc. International Symposium on Nonlinear Theory and Its Application, pp. 923–926 (1998)
Nishikawa, T., Saruwatari, H., Shikano, K.: Blind source separation of acoustic signals based on multistage ICA combining frequency-domain ICA and time-domain ICA. IEICE Trans. Fundamentals E86-A, 846–858 (2003)
Araki, S., Mukai, R., Makino, S., Nishikawa, T., Saruwatari, H.: The fundamental limitation of frequency domain blind source separation for convolutive mixtures of speech. IEEE Trans. Speech and Audio Processing. 11, 109–116 (2003)
Choi, S., Amari, S., Cichocki, A., Liu, R.: Natural gradient learning with a nonholonomic constraint for blind deconvolution of multiple channels. In: Proc. International Workshop on ICA and BSS, pp. 371–376 (1999)
Nishikawa, T., Saruwatari, H., Shikano, K.: Stable learning algorithm for blind separation of temporally correlated acoustic signals combining multistage ICA and Linear Prediction. IEICE Trans. Fundamentals E86-A, 2028–2036 (2003)
Nishikawa, T., Abe, H., Saruwatari, H., Shikano, K.: Overdetermined Blind Separation for Convolutive Mixtures of Speech Based on Multistage ICA using Subarray Processing In: Proc. ICASSP 2004 (2004) (accepted)
Nakamura, S., Hiyane, K., Asano, F., Nishiura, T., Yamada, T.: Acoustical sound database in real environments for sound scene understanding and hands-free speech recognition In: Proc. International Conference on Language Resources and Evaluation, pp. 965–968 (2000)
Aichner, R., Araki, S., Makino, S., Nishikawa, T., Saruwatari, H.: Time domain ICA blind source separation of non-stationary convolved signals by utilizing geometric beamforming In: Proc. IEEE International Workshop on Neural Networks for Signal Processing, pp. 445–454 (2002)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2004 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Nishikawa, T., Saruwatari, H., Shikano, K., Kaminuma, A. (2004). Stable and Low-Distortion Algorithm Based on Overdetermined Blind Separation for Convolutive Mixtures of Speech. In: Puntonet, C.G., Prieto, A. (eds) Independent Component Analysis and Blind Signal Separation. ICA 2004. Lecture Notes in Computer Science, vol 3195. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-30110-3_111
Download citation
DOI: https://doi.org/10.1007/978-3-540-30110-3_111
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-23056-4
Online ISBN: 978-3-540-30110-3
eBook Packages: Springer Book Archive