Blind Source Separation Using Mixtures of Alpha-Stable Distributions
This addresses the problem of separating audio sources in convolutive mixes for applications like speech enhancement, though it appears incremental as it builds on existing alpha-stable distribution models with a new inference method.
The authors tackled the problem of blind source separation for audio signals by developing a new algorithm based on mixtures of alpha-stable distributions, which better model audio in the time-frequency domain than Gaussian distributions. They showed that this method yields better separation performance than Gaussian-based binary-masking methods.
We propose a new blind source separation algorithm based on mixtures of alpha-stable distributions. Complex symmetric alpha-stable distributions have been recently showed to better model audio signals in the time-frequency domain than classical Gaussian distributions thanks to their larger dynamic range. However, inference of these models is notoriously hard to perform because their probability density functions do not have a closed-form expression in general. Here, we introduce a novel method for estimating mixture of alpha-stable distributions based on characteristic function matching. We apply this to the blind estimation of binary masks in individual frequency bands from multichannel convolutive audio mixes. We show that the proposed method yields better separation performance than Gaussian-based binary-masking methods.