Single Channel Audio Source Separation using Deep Neural Network Ensembles

Grais, Emad M, Roma, G, Simpson, AJR and Plumbley, Mark (2016) Single Channel Audio Source Separation using Deep Neural Network Ensembles In: 140th Convention of the Audio Engineering Society, 2016-06-04 - 2016-06-07, Paris, France.

Preview	Text Grais_aes140.pdf - Accepted version Manuscript Download (140kB) | Preview
Preview	Text (licence) SRI_deposit_agreement.pdf Available under License : See the attached licence file. Download (33kB) | Preview
	Text grais_aes140.pdf - Version of Record Restricted to Repository staff only Available under License : See the attached licence file. Download (143kB)

Abstract

Deep neural networks (DNNs) are often used to tackle the single channel source separation (SCSS) problem by predicting time-frequency masks. The predicted masks are then used to separate the sources from the mixed signal. Different types of masks produce separated sources with different levels of distortion and interference. Some types of masks produce separated sources with low distortion, while other masks produce low interference between the separated sources. In this paper, a combination of different DNNs’ predictions (masks) is used for SCSS to achieve better quality of the separated sources than using each DNN individually. We train four different DNNs by minimizing four different cost functions to predict four different masks. The first and second DNNs are trained to approximate reference binary and soft masks. The third DNN is trained to predict a mask from the reference sources directly. The last DNN is trained similarly to the third DNN but with an additional discriminative constraint to maximize the differences between the estimated sources. Our experimental results show that combining the predictions of different DNNs achieves separated sources with better quality than using each DNN individually

Item Type:	Conference or Workshop Item (Conference Paper)
Subjects :	Electronic Engineering
Divisions :	Faculty of Engineering and Physical Sciences > Electronic Engineering
Authors :	Name Email ORCID Grais, Emad M grais@surrey.ac.uk Roma, G Simpson, AJR Plumbley, Mark m.plumbley@surrey.ac.uk
Date :	26 May 2016
Copyright Disclaimer :	Copyright 2016 Audio Engineering Society. This paper was presented at the 140th Convention of the Audio Engineering Society, as paper number 9494. The full published version can be found at http://www.aes.org/e-lib/browse.cfm?elib=18193.
Contributors :	Contribution Name Email ORCID AES,
Related URLs :	Publisher
Depositing User :	Symplectic Elements
Date Deposited :	08 Jul 2016 17:11
Last Modified :	16 Jan 2019 17:05
URI:	http://epubs.surrey.ac.uk/id/eprint/811154

Actions (login required)

View Item

Downloads