Validity of the Independence Assumption for the Separation of Instantaneous and Convolutive Mixtures of Speech and Music Sources

Abstract

In this paper, we study the validity of the assumption that speech source signals exhibit lower dependency and therefore better separability with Independent Component Analysis algorithms than music sources. In particular, we investigate some dependency measures in the temporal and the time-frequency domains, resp. in the framework of instantaneous and convolutive mixtures. Moreover, we test several ICA methods, based on the above dependency measures, on the same source signals. We experimentally show that speech and music sources tend to have the same mean behaviour for excerpt durations above 20 ms, but music signals provide more spread dependency measures and SIR values. Lastly, we experimentally show that Gaussian nonstationary mutual information is better suited to audio signals than mutual information.