Under-Determined Reverberant Audio Source Separation Using Local Observed Covariance and Auditory-Motivated Time-Frequency Representation

Ngoc Duong 1 Emmanuel Vincent 1 Rémi Gribonval 1
1 METISS - Speech and sound data modeling and processing
IRISA - Institut de Recherche en Informatique et Systèmes Aléatoires, Inria Rennes – Bretagne Atlantique
Abstract : We consider the local Gaussian modeling framework for under-determined convolutive audio source separation, where the spatial image of each source is modeled as a zero-mean Gaussian variable with full-rank time- and frequency- dependent covariance. We investigate two methods to improve the accuracy of parameter estimation, based on the use of local observed covariance and auditory-motivated time-frequency representation. We derive an iterative expectation-maximization (EM) algorithm with a suitable initialization scheme. Experimental results over stereo synthetic reverberant mixtures of speech show the effectiveness of the proposed methods.
Complete list of metadatas

Cited literature [13 references]  Display  Hide  Download

https://hal.inria.fr/inria-00541868
Contributor : Rémi Gribonval <>
Submitted on : Saturday, February 5, 2011 - 1:00:11 PM
Last modification on : Thursday, March 21, 2019 - 2:20:42 PM
Long-term archiving on : Friday, May 6, 2011 - 2:25:49 AM

File

2010_LVA_DuongEtAl.pdf
Files produced by the author(s)

Identifiers

Citation

Ngoc Duong, Emmanuel Vincent, Rémi Gribonval. Under-Determined Reverberant Audio Source Separation Using Local Observed Covariance and Auditory-Motivated Time-Frequency Representation. Latent Variable Analysis and Signal Separation, 9th International Conference on, Sep 2010, Saint-Malo, France. pp.73--80, ⟨10.1007/978-3-642-15995-4_10⟩. ⟨inria-00541868⟩

Share

Metrics

Record views

638

Files downloads

310