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. © Springer-Verlag Berlin Heidelberg 2007.
CITATION STYLE
Vincent, E. (2007). Complex nonconvex lp norm minimization for underdetermined source separation. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 4666 LNCS, pp. 430–437). Springer Verlag. https://doi.org/10.1007/978-3-540-74494-8_54
Mendeley helps you to discover research relevant for your work.