HAL will be down for maintenance from Friday, June 10 at 4pm through Monday, June 13 at 9am. More information
Skip to Main content Skip to Navigation
Journal articles

Under-determined reverberant audio source separation using a full-rank spatial covariance model

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 : This article addresses the modeling of reverberant recording environments in the context of under-determined convolutive blind source separation. We model the contribution of each source to all mixture channels in the time-frequency domain as a zero-mean Gaussian random variable whose covari- ance encodes the spatial characteristics of the source. We then consider four specific covariance models, including a full-rank unconstrained model. We derive a family of iterative expectation- maximization (EM) algorithms to estimate the parameters of each model and propose suitable procedures adapted from the state- of-the-art to initialize the parameters and to align the order of the estimated sources across all frequency bins. Experimental results over reverberant synthetic mixtures and live recordings of speech data show the effectiveness of the proposed approach.
Complete list of metadata

Cited literature [24 references]  Display  Hide  Download

Contributor : Rémi Gribonval Connect in order to contact the contributor
Submitted on : Thursday, January 27, 2011 - 9:48:52 PM
Last modification on : Friday, February 4, 2022 - 3:25:18 AM
Long-term archiving on: : Thursday, April 28, 2011 - 2:30:59 AM


Files produced by the author(s)



Ngoc Duong, Emmanuel Vincent, Rémi Gribonval. Under-determined reverberant audio source separation using a full-rank spatial covariance model. IEEE Transactions on Audio, Speech and Language Processing, Institute of Electrical and Electronics Engineers, 2010, 18 (7), pp.1830--1840. ⟨10.1109/TASL.2010.2050716⟩. ⟨inria-00541865⟩



Record views


Files downloads