INTERSPEECH 2011

This paper presents an efficient algorithm to solve Lpnorm minimization problem for underdetermined speech separation; that is, for the case that there are more sound sources than microphones. We employ an auxiliary function method in order to derive update rules under the assumption that the amplitude of each sound source follows generalized Gaussian distribution. Experiments reveal that our method solves the L1norm minimization problem ten times faster than a general solver, and also solves Lpnorm minimization problem efficiently, especially when the parameter p is small; when p is not more than 0.7, it runs in realtime without loss of separation quality.
Bibliographic reference. Hirasawa, Yasuharu / Yasuraoka, Naoki / Takahashi, Toru / Ogata, Tetsuya / Okuno, Hiroshi G. (2011): "Fast and simple iterative algorithm of lpnorm minimization for underdetermined speech separation", In INTERSPEECH2011, 17451748.