Speech dereverberation performance of a polynomial-EVD subspace approach
Speech dereverberation performance of a polynomial-EVD subspace approach
The degradation of speech arising from additive background noise and reverberation affects the performance of important speech applications such as telecommunications, hearing aids, voice-controlled systems and robot audition. In this work, we focus on dereverberation. It is shown that the parameterized polynomial matrix eigenvalue decomposition (PEVD)-based speech enhancement algorithm exploits the lack of correlation between speech and the late reflections to enhance the speech component associated with the direct path and early reflections. The algorithm's performance is evaluated using simulations involving measured acoustic impulse responses and noise from the ACE corpus. The simulations and informal listening examples have indicated that the PEVD-based algorithm performs dereverberation over a range of SNRs without introducing any noticeable processing artefacts.
Broadband signal processing, Convolutive noise, Dereverberation, Microphone array, Polynomial matrix eigenvalue decomposition
221-225
European Signal Processing Conference
Neo, Vincent W.
7ec5cc5f-8248-40ec-8864-b31335d4ddf2
Evers, Christine
93090c84-e984-4cc3-9363-fbf3f3639c4b
Naylor, Patrick A.
13079486-664a-414c-a1a2-01a30bf0997b
24 January 2021
Neo, Vincent W.
7ec5cc5f-8248-40ec-8864-b31335d4ddf2
Evers, Christine
93090c84-e984-4cc3-9363-fbf3f3639c4b
Naylor, Patrick A.
13079486-664a-414c-a1a2-01a30bf0997b
Neo, Vincent W., Evers, Christine and Naylor, Patrick A.
(2021)
Speech dereverberation performance of a polynomial-EVD subspace approach.
In 28th European Signal Processing Conference, EUSIPCO 2020 - Proceedings.
vol. 2021-January,
European Signal Processing Conference.
.
(doi:10.23919/Eusipco47968.2020.9287869).
Record type:
Conference or Workshop Item
(Paper)
Abstract
The degradation of speech arising from additive background noise and reverberation affects the performance of important speech applications such as telecommunications, hearing aids, voice-controlled systems and robot audition. In this work, we focus on dereverberation. It is shown that the parameterized polynomial matrix eigenvalue decomposition (PEVD)-based speech enhancement algorithm exploits the lack of correlation between speech and the late reflections to enhance the speech component associated with the direct path and early reflections. The algorithm's performance is evaluated using simulations involving measured acoustic impulse responses and noise from the ACE corpus. The simulations and informal listening examples have indicated that the PEVD-based algorithm performs dereverberation over a range of SNRs without introducing any noticeable processing artefacts.
This record has no associated files available for download.
More information
Published date: 24 January 2021
Additional Information:
Funding Information:
† The research leading to these results is funded through the UK EPSRC Fellowship grant no. EP/P001017/1 while the author was with the Department of Electrical and Electronic Engineering, Imperial College London. ∗ This work is funded through the UK EPSRC grant no. EP/S035842/1.
Publisher Copyright:
© 2021 European Signal Processing Conference, EUSIPCO. All rights reserved.
Copyright:
Copyright 2021 Elsevier B.V., All rights reserved.
Venue - Dates:
28th European Signal Processing Conference, EUSIPCO 2020, , Amsterdam, Netherlands, 2020-08-24 - 2020-08-28
Keywords:
Broadband signal processing, Convolutive noise, Dereverberation, Microphone array, Polynomial matrix eigenvalue decomposition
Identifiers
Local EPrints ID: 450390
URI: http://eprints.soton.ac.uk/id/eprint/450390
ISSN: 2219-5491
PURE UUID: 55a35c71-65e4-448d-b2da-4360b97d33e1
Catalogue record
Date deposited: 27 Jul 2021 16:30
Last modified: 18 Mar 2024 03:56
Export record
Altmetrics
Contributors
Author:
Vincent W. Neo
Author:
Christine Evers
Author:
Patrick A. Naylor
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics