Non negative sparse representation for Wiener based source separation with a single sensor

Author

Benaroya, Laurent ; Donagh, Lorcan M. ; Bimbot, Frédéric ; Gribonval, Rémi

Author_Institution

IRISA, Rennes, France

Volume

6

fYear

2003

fDate

6-10 April 2003

Abstract

We propose a new method to perform the separation of two sound sources from a single sensor. This method generalizes Wiener filtering with locally stationary, non-Gaussian, parametric source models. The method involves a learning phase for which we propose three different algorithm. In the separation phase, we use a sparse non negative decomposition algorithm of our own. The algorithms are evaluated on the separation of real audio data.

Keywords

Wiener filters; audio signal processing; learning (artificial intelligence); parameter estimation; signal representation; source separation; Wiener filtering; learning phase; nonGaussian source models; nonnegative decomposition algorithm; nonnegative sparse representation; optimal estimates; parametric source models; real audio data; separation phase; sound source separation; sparse decomposition algorithm; stationary source models; Acoustic sensors; Fourier transforms; Frequency estimation; Gaussian processes; Parameter estimation; Parametric statistics; Phase estimation; Source separation; Spectral shape; Wiener filter;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech, and Signal Processing, 2003. Proceedings. (ICASSP '03). 2003 IEEE International Conference on

ISSN

1520-6149

Print_ISBN

0-7803-7663-3

Type

conf

DOI

10.1109/ICASSP.2003.1201756

Filename

1201756