Underdetermined separation of instantaneous speech mixtures with hierarchical sparse recovery scheme

This paper describes a novel algorithm for underdetermined instantaneous speech separation problem based on hierarchical sparse Bayesian technique for efficient data reconstruction. The proposed algorithm consists of three steps. The unknown mixing matrix is firstly estimated from the speech mixtures in the transform domain. Then, a permutation issue is solved based on the results from the first step to get the correct order of the dictionary. Finally speech sources are recovered using the hierarchical sparse structure of the mixed speech signals. Numerical experiments including the comparison with other sparse representation approach are provided to show that our proposed method could reduce the interference effectively and achieve desirable performance improvement.