Combining time-delayed decorrelation and ICA: towards solving the cocktail party problem

We present methods to separate blindly mixed signals recorded in a room. The learning algorithm is based on the information maximization in a single layer neural network. We focus on the implementation of the learning algorithm and on issues that arise when separating speakers in room recordings. We used an infomax approach in a feedforward neural network implemented in the frequency domain using the polynomial filter matrix algebra technique. A fast convergence speed was achieved by using a time-delayed decorrelation method as a preprocessing step. Under minimum-phase mixing conditions this preprocessing step was sufficient for the separation of signals. These methods successfully separated a recorded voice with music in the background (cocktail party problem). Finally, we discuss problems that arise in real world recordings and their potential solutions.

[1]  E. Ferrara.  Fast implementations of LMS adaptive filters , 1980 .

[2]  R. Lambert.  Multichannel blind deconvolution: FIR matrix algebra and separation of multipath mixtures , 1996 .

[3]  Terrence J. Sejnowski,et al.  An Information-Maximization Approach to Blind Separation and Blind Deconvolution , 1995, Neural Computation.

[4]  Reinhold Orglmeister,et al.  Blind source separation of real world signals , 1997, Proceedings of International Conference on Neural Networks (ICNN'97).

[5]  Jean-François Cardoso,et al.  Equivariant adaptive source separation , 1996, IEEE Trans. Signal Process..

[6]  Te-Won Lee,et al.  Blind Separation of Delayed and Convolved Sources , 1996, NIPS.

[7]  Ehud Weinstein,et al.  Multichannel signal separation: methods and analysis , 1996, IEEE Trans. Signal Process..

[8]  Colin Fyfe,et al.  An extended exploratory projection pursuit network with linear and nonlinear anti-hebbian lateral connections applied to the cocktail party problem , 1997, Neural Networks.

[9]  S. Amari,et al.  Multichannel blind separation and deconvolution of sources with arbitrary distributions , 1997, Neural Networks for Signal Processing VII. Proceedings of the 1997 IEEE Signal Processing Society Workshop.

[10]  Shun-ichi Amari,et al.  Neural Network Models for Blind Separation of Time Delayed and Convolved Signals , 1997 .

[11]  Kari Torkkola,et al.  Blind separation of convolved sources based on information maximization , 1996, Neural Networks for Signal Processing VI. Proceedings of the 1996 IEEE Signal Processing Society Workshop.

[12]  H. G. Schuster,et al.  Blind separation of convolutive mixtures , 1997, Proceedings of 40th Midwest Symposium on Circuits and Systems. Dedicated to the Memory of Professor Mac Van Valkenburg.

[13]  Schuster,et al.  Separation of a mixture of independent signals using time delayed correlations. , 1994, Physical review letters.

[14]  Shun-ichi Amari,et al.  Natural Gradient Works Efficiently in Learning , 1998, Neural Computation.

[15]  S. Haykin,et al.  Adaptive Filter Theory , 1986 .