Time-Domain Blind Audio Source Separation using Advanced ICA Methods

In this paper, a prototype of novel algorithm for blind sepa-ration of convolutive mixtures of audio sources is proposed. The method works in time-domain, and it is based on the re-cently very successful algorithm EFICA for Independent Com-ponent Analysis, which is an enhanced version of more famous FastICA. Performance of the new algorithm is very promising, at least, comparable to other (mostly frequency domain) algo-rithms. Audio separation examples are included. Index Terms: independent component analysis, multichannel blind deconvolution, audio source separatio