Multichannel Audio Modeling with Elliptically Stable Tensor Decomposition - Archive ouverte HAL Access content directly
Conference Papers Year : 2018

Multichannel Audio Modeling with Elliptically Stable Tensor Decomposition


This paper introduces a new method for multichannel speech enhancement based on a versatile modeling of the residual noise spec-trogram. Such a model has already been presented before in the single channel case where the noise component is assumed to follow an alpha-stable distribution for each time-frequency bin, whereas the speech spec-trogram, supposed to be more regular, is modeled as Gaussian. In this paper, we describe a multichannel extension of this model, as well as a Monte Carlo Expectation-Maximisation algorithm for parameter estimation. In particular, a multichannel extension of the Itakura-Saito nonnegative matrix factorization is exploited to estimate the spectral parameters for speech, and a Metropolis-Hastings algorithm is proposed to estimate the noise contribution. We evaluate the proposed method in a challenging multichannel denoising application and compare it to other state-of-the-art algorithms.
Fichier principal
Vignette du fichier
LVA-ICA2018_046_original_v5.pdf (366.42 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

lirmm-01766795 , version 1 (14-04-2018)



Mathieu Fontaine, Fabian-Robert Stöter, Antoine Liutkus, Umut Şimşekli, Romain Serizel, et al.. Multichannel Audio Modeling with Elliptically Stable Tensor Decomposition. 14th International Conference on Latent Variable Analysis and Signal Separation (LVA/ICA 2018), Jul 2018, Surrey, United Kingdom. pp.13-23, ⟨10.1007/978-3-319-93764-9_2⟩. ⟨lirmm-01766795⟩
612 View
533 Download



Gmail Facebook Twitter LinkedIn More