University of Surrey

Test tubes in the lab Research in the ATI Dance Research

Audio source separation with deep neural networks using the dropout algorithm

Zermini, Alfredo, Wang, Wenwu, Kong, Qiuqiang, Xu, Yong and Plumbley, Mark (2017) Audio source separation with deep neural networks using the dropout algorithm In: Signal Processing with Adaptive Sparse Structured Representations (SPARS) 2017, 05-08 Jun 2017, Instituto Superior Técnico, University of Lisbon, Lisbon, Portugal.

[img]
Preview
Text
Audio source separation with deep neural networks using the dropout algorithm.pdf - Accepted version Manuscript

Download (1MB) | Preview

Abstract

A method based on Deep Neural Networks (DNNs) and time-frequency masking has been recently developed for binaural audio source separation. In this method, the DNNs are used to predict the Direction Of Arrival (DOA) of the audio sources with respect to the listener which is then used to generate soft time-frequency masks for the recovery/estimation of the individual audio sources. In this paper, an algorithm called ‘dropout’ will be applied to the hidden layers, affecting the sparsity of hidden units activations: randomly selected neurons and their connections are dropped during the training phase, preventing feature co-adaptation. These methods are evaluated on binaural mixtures generated with Binaural Room Impulse Responses (BRIRs), accounting a certain level of room reverberation. The results show that the proposed DNNs system with randomly deleted neurons is able to achieve higher SDRs performances compared to the baseline method without the dropout algorithm.

Item Type: Conference or Workshop Item (Conference Paper)
Divisions : Faculty of Engineering and Physical Sciences > Electronic Engineering
Authors :
NameEmailORCID
Zermini, Alfredoalfredo.zermini@surrey.ac.ukUNSPECIFIED
Wang, WenwuW.Wang@surrey.ac.ukUNSPECIFIED
Kong, Qiuqiangq.kong@surrey.ac.ukUNSPECIFIED
Xu, Yongyong.xu@surrey.ac.ukUNSPECIFIED
Plumbley, Markm.plumbley@surrey.ac.ukUNSPECIFIED
Date : 8 June 2017
Copyright Disclaimer : © 2017 The Authors.
Related URLs :
Depositing User : Clive Harris
Date Deposited : 11 Aug 2017 07:44
Last Modified : 11 Aug 2017 07:44
URI: http://epubs.surrey.ac.uk/id/eprint/841888

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year


Information about this web site

© The University of Surrey, Guildford, Surrey, GU2 7XH, United Kingdom.
+44 (0)1483 300800