Abstract
This paper presents a fully Bayesian hierarchical model for blind audio source separation in a noisy environment. Our probabilistic approach is based on Gaussian priors for the speech signals, Gamma hyperpriors for the speech precisions and a Gamma prior for the noise precision. The time-varying acoustic channels are modelled with a linear-Gaussian state-space model. The inference is carried out using a variational Expectation-Maximization (VEM) algorithm, leading to a variant of the multi-speaker multichannel Wiener filter (MCWF) to separate and enhance the audio sources, and a Kalman smoother to infer the acoustic channels. The VEM speech estimator can be decomposed into two stages: A multi-speaker linearly constrained minimum variance (LCMV) beamformer followed by a variational multi-speaker postfilter. The proposed algorithm is evaluated in a static scenario using recorded room impulse responses (RIRs) with two reverberation levels, showing superior performance compared to competing methods.
Original language | English |
---|---|
Title of host publication | 28th European Signal Processing Conference, EUSIPCO 2020 - Proceedings |
Publisher | European Signal Processing Conference, EUSIPCO |
Pages | 276-280 |
Number of pages | 5 |
ISBN (Electronic) | 9789082797053 |
DOIs | |
State | Published - 24 Jan 2021 |
Event | 28th European Signal Processing Conference, EUSIPCO 2020 - Amsterdam, Netherlands Duration: 24 Aug 2020 → 28 Aug 2020 |
Publication series
Name | European Signal Processing Conference |
---|---|
Volume | 2021-January |
ISSN (Print) | 2219-5491 |
Conference
Conference | 28th European Signal Processing Conference, EUSIPCO 2020 |
---|---|
Country/Territory | Netherlands |
City | Amsterdam |
Period | 24/08/20 → 28/08/20 |
Bibliographical note
Publisher Copyright:© 2021 European Signal Processing Conference, EUSIPCO. All rights reserved.
Funding
This project has received funding from the European Union’s Horizon 2020 Research and Innovation Programme, Grant Agreement No. 871245. Yaron Laufer is supported by the Adams Fellowship Program of the Israel Academy of Sciences and Humanities. This project has received funding from the European Union's Horizon 2020 Research and Innovation Programme, Grant Agreement No. 871245. Yaron Laufer is supported by the Adams Fellowship Program of the Israel Academy of Sciences and Humanities.
Funders | Funder number |
---|---|
Horizon 2020 Framework Programme | |
Israel Academy of Sciences and Humanities | |
Horizon 2020 | 871245 |
Keywords
- Audio source separation
- Variational EM