Hierarchical clustering applied to overcomplete BSS for convolutive mixtures

In this paper we address the problem of overcomplete BSS for convolutive mixtures following a two-step approach. In the first step the mixing matrix is estimated, which is then used to separate the signals in the second step. For estimating the mixing matrix we propose an algorithm based on hierarchical clustering, assuming that the source signals are sufficiently sparse. It has the advantage of working directly on the complex valued sample data in the frequency-domain. It also shows better convergence than algorithms based on selforganizing maps. The results are improved by reducing the variance of direction of arrival. Experiments show accurate estimations of the mixing matrix and very low musical tone noise even in reverberant environment.

[1]  Hiroshi Sawada,et al.  A robust and precise method for solving the permutation problem of frequency-domain blind source separation , 2004, IEEE Transactions on Speech and Audio Processing.

[2]  Fionn Murtagh,et al.  Comments on 'Parallel Algorithms for Hierarchical Clustering and Cluster Validity' , 1992, IEEE Trans. Pattern Anal. Mach. Intell..

[3]  L. Vielva,et al.  UNDERDETERMINED BLIND SOURCE SEPARATION USING A PROBABILISTIC SOURCE SPARSITY MODEL , 2001 .

[4]  Scott Rickard,et al.  Blind separation of speech mixtures via time-frequency masking , 2004, IEEE Transactions on Signal Processing.

[5]  Deniz Erdogmus,et al.  Estimation of the mixing matrix for underdetermined blind source separation using spectral estimation techniques , 2002, 2002 11th European Signal Processing Conference.

[6]  Hiroshi Sawada,et al.  Blind Separation of More Speech than Sensors with Less Distortion by Combining Sparseness and ICA , 2003 .

[7]  K. Matsuoka,et al.  Independent Component Analysis and Its Applications to Sound Signal Separation , 2003 .

[8]  Fabian J. Theis,et al.  Mathematics in independent component analysis , 2003, Seventh International Symposium on Signal Processing and Its Applications, 2003. Proceedings..

[9]  Hiroshi Sawada,et al.  GEOMETRICAL INTERPRETATION OF THE PCA SUBSPACE METHOD FOR OVERDETERMINED BLIND SOURCE SEPARATION , 2003 .

[10]  D. Ruppert.  The Elements of Statistical Learning: Data Mining, Inference, and Prediction , 2004 .

[11]  Shoko Araki,et al.  Blind Source Separation when Speech Signals Outnumber Sensors using a Sparseness - Mixing Matrix Estimation (SMME) , 2003 .

[12]  Pau Bofill,et al.  Underdetermined blind separation of delayed sound sources in the frequency domain , 2003, Neurocomputing.

[13]  Fathi M. Salem,et al.  ALGEBRAIC OVERCOMPLETE INDEPENDENT COMPONENT ANALYSIS , 2003 .

[14]  Özgür Yilmaz,et al.  On the approximate W-disjoint orthogonality of speech , 2002, 2002 IEEE International Conference on Acoustics, Speech, and Signal Processing.