ICA-based clustering for resolving permutation ambiguity in frequency-domain convolutive source separation

Permutation ambiguity is an inherent limitation in independent component analysis, which is a bottleneck in frequency-domain methods of convolutive source separation. In this paper we present a method for resolving this permutation ambiguity, where we group vectors of estimated frequency responses into clusters in such a way that each cluster contains frequency responses associated with the same source. The clustering is carried out, applying independent component analysis to estimated frequency responses. In contrast to existing methods, the proposed method does not require any prior information such as the geometric configuration of microphone arrays or distances between sources and microphones. Experimental results confirm the validity of our method.