Online Single-Microphone Source Separation using Non-Linear Autoregressive Models

Bart van Erp, Bert de Vries
Proceedings of The 11th International Conference on Probabilistic Graphical Models, PMLR 186:37-48, 2022.

Abstract

In this paper a modular approach to single-microphone source separation is proposed. A probabilistic model for mixtures of observations is constructed, where the independent underlying source signals are described by non-linear autoregressive models. Source separation in this model is achieved by performing online probabilistic inference through an efficient message passing procedure. For retaining tractability with the non-linear autoregressive models, three different approximation methods are described. A set of experiments shows the effectiveness of the proposed source separation approach. The source separation performance of the different approximation methods is quantified through a set of verification experiments. Our approach is validated in a speech denoising task.

Cite this Paper

BibTeX
@InProceedings{pmlr-v186-erp22a, title = {Online Single-Microphone Source Separation using Non-Linear Autoregressive Models}, author = {van Erp, Bart and de Vries, Bert}, booktitle = {Proceedings of The 11th International Conference on Probabilistic Graphical Models}, pages = {37--48}, year = {2022}, editor = {Salmerón, Antonio and Rumı́, Rafael}, volume = {186}, series = {Proceedings of Machine Learning Research}, month = {05--07 Oct}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v186/erp22a/erp22a.pdf}, url = {https://proceedings.mlr.press/v186/erp22a.html}, abstract = {In this paper a modular approach to single-microphone source separation is proposed. A probabilistic model for mixtures of observations is constructed, where the independent underlying source signals are described by non-linear autoregressive models. Source separation in this model is achieved by performing online probabilistic inference through an efficient message passing procedure. For retaining tractability with the non-linear autoregressive models, three different approximation methods are described. A set of experiments shows the effectiveness of the proposed source separation approach. The source separation performance of the different approximation methods is quantified through a set of verification experiments. Our approach is validated in a speech denoising task.} }
Endnote
%0 Conference Paper %T Online Single-Microphone Source Separation using Non-Linear Autoregressive Models %A Bart van Erp %A Bert de Vries %B Proceedings of The 11th International Conference on Probabilistic Graphical Models %C Proceedings of Machine Learning Research %D 2022 %E Antonio Salmerón %E Rafael Rumı́ %F pmlr-v186-erp22a %I PMLR %P 37--48 %U https://proceedings.mlr.press/v186/erp22a.html %V 186 %X In this paper a modular approach to single-microphone source separation is proposed. A probabilistic model for mixtures of observations is constructed, where the independent underlying source signals are described by non-linear autoregressive models. Source separation in this model is achieved by performing online probabilistic inference through an efficient message passing procedure. For retaining tractability with the non-linear autoregressive models, three different approximation methods are described. A set of experiments shows the effectiveness of the proposed source separation approach. The source separation performance of the different approximation methods is quantified through a set of verification experiments. Our approach is validated in a speech denoising task.
APA
van Erp, B. & de Vries, B.. (2022). Online Single-Microphone Source Separation using Non-Linear Autoregressive Models. Proceedings of The 11th International Conference on Probabilistic Graphical Models, in Proceedings of Machine Learning Research 186:37-48 Available from https://proceedings.mlr.press/v186/erp22a.html.

Related Material