Complex nonconvex lp norm minimization for underdetermined source separation

Emmanuel Vincent (1)

1 METISS - Speech and sound data modeling and processing

Underdetermined source separation methods often rely on the assumption that the time-frequency source coefficients are independent and Laplacian distributed. In this article, we extend these methods by assuming that these coefficients follow a generalized Gaussian prior with shape parameter p. We study mathematical and experimental properties of the resulting complex nonconvex lp norm optimization problem in a particular case and derive an efficient global optimization algorithm. We show that the best separation performance for three-source stereo convolutive speech mixtures is achieved for small p.

Domaines: Traitement du signal et de l'image [eess.SP]; Traitement du signal et de l'image [eess.SP]

Emmanuel Vincent. Complex nonconvex lp norm minimization for underdetermined source separation. 7th Int. Conf. on Independent Component Analysis and Blind Source Separation (ICA), Sep 2007, London, United Kingdom. pp.430--437. ⟨inria-00544203⟩