Spatial location priors for Gaussian model based reverberant audio source separation - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Reports (Research Report) Year : 2012

Spatial location priors for Gaussian model based reverberant audio source separation

Abstract

We consider the Gaussian framework for reverberant audio source separation, where the sources are modeled in the time-frequency domain by their short-term power spectra and their spatial covariance matrices. We propose three alternative probabilistic priors over the spatial covariance matrices which are consistent with the theory of statistical room acoustics and we derive Expectation-Maximization (EM) algorithms for maximum a posteriori (MAP) estimation. We argue that these algorithms provide a statistically principled solution to the permutation problem and to the risk of overfitting resulting from conventional maximum likelihood (ML) estimation. We show experimentally that, in a semi-informed scenario where the source positions and certain room characteristics are known, the algorithms using respectively inverse-Wishart and Gaussian priors outperform their ML counterparts. This opens the way to rigorous statistical treatment of this family of models in other scenarios in the future.
Fichier principal
Vignette du fichier
RR-8057.pdf (268.69 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-00727781 , version 1 (04-09-2012)
hal-00727781 , version 2 (02-04-2013)

Identifiers

  • HAL Id : hal-00727781 , version 2

Cite

Ngoc Q. K. Duong, Emmanuel Vincent, Rémi Gribonval. Spatial location priors for Gaussian model based reverberant audio source separation. [Research Report] RR-8057, INRIA. 2012. ⟨hal-00727781v2⟩
457 View
436 Download

Share

Gmail Facebook X LinkedIn More