On-line non-stationary ICA using mixture models

In this paper we address the problem of on-line source separation with sources modelled as mixtures of Gaussians which are linearly combined via a series of non-stationary mixing matrices. The online recovery of the sources from the observations is a non-linear statistical filtering problem that we address using state of the art particle filter methods. Simulations are presented and satisfactory results are obtained.

[1]  R. Kohn,et al.  On Gibbs sampling for state space models , 1994 .

[2]  Christophe Andrieu,et al.  Iterative algorithms for optimal state estimation of jump Markov linear systems , 1999, 1999 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings. ICASSP99 (Cat. No.99CH36258).

[3]  Pierre Del Moral,et al.  Discrete Filtering Using Branching and Interacting Particle Systems , 1998 .

[4]  Douglas A. Reynolds,et al.  Robust text-independent speaker identification using Gaussian mixture speaker models , 1995, IEEE Trans. Speech Audio Process..

[5]  Xiao Yu,et al.  Speech separation based on the GMM PDF estimation , 1998, ICSLP.

[6]  P. Green,et al.  On Bayesian Analysis of Mixtures with an Unknown Number of Components (with discussion) , 1997 .

[7]  Peter E. Hart,et al.  Pattern classification and scene analysis , 1974, A Wiley-Interscience publication.

[8]  Eric Moulines,et al.  Maximum likelihood for blind separation and deconvolution of noisy signals using mixture models , 1997, 1997 IEEE International Conference on Acoustics, Speech, and Signal Processing.