TY - GEN
T1 - Multichannel audio modeling with elliptically stable tensor decomposition
AU - Fontaine, Mathieu
AU - Stöter, Fabian Robert
AU - Liutkus, Antoine
AU - Şimşekli, Umut
AU - Serizel, Romain
AU - Badeau, Roland
N1 - Publisher Copyright:
© Springer International Publishing AG, part of Springer Nature 2018.
PY - 2018/1/1
Y1 - 2018/1/1
N2 - This paper introduces a new method for multichannel speech enhancement based on a versatile modeling of the residual noise spectrogram. 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 spectrogram, 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.
AB - This paper introduces a new method for multichannel speech enhancement based on a versatile modeling of the residual noise spectrogram. 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 spectrogram, 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.
UR - https://www.scopus.com/pages/publications/85048583815
U2 - 10.1007/978-3-319-93764-9_2
DO - 10.1007/978-3-319-93764-9_2
M3 - Conference contribution
AN - SCOPUS:85048583815
SN - 9783319937632
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 13
EP - 23
BT - Latent Variable Analysis and Signal Separation - 14th International Conference, LVA/ICA 2018, Proceedings
A2 - Gannot, Sharon
A2 - Deville, Yannick
A2 - Mason, Russell
A2 - Plumbley, Mark D.
A2 - Ward, Dominic
PB - Springer Verlag
T2 - 14th International Conference on Latent Variable Analysis and Signal Separation, LVA/ICA 2018
Y2 - 2 July 2018 through 5 July 2018
ER -