Acoustic source separation has received a great interest from the research community but the problem remains challenging in the undetermined setting (when the number of sources I is less than the number of channels J), including the single-channel case (I = 1), and for convolutive mixtures. It is now quite clear that source separation performances strongly depend on the amount of available prior information about the sources and the mixing process one can introduce in the source separation algorithm. The so-called informed source separation methods then appear to be particularly attractive.
This special issue will also build on the success of the recent (double) special session at Eusipco 2012 on the very same topic.
Edited by: Ozerov Alexey, Ali Taylan Cemgil, Derry FitzGerald, Gaël Richard and Tuomas Virtanen