This is the latest version of this eprint.
Close, G. orcid.org/0000-0002-9478-5421, Hain, T. orcid.org/0000-0003-0939-3464 and Goetze, S. orcid.org/0000-0003-1044-7343 (2022) MetricGAN+/-: increasing robustness of noise reduction on unseen data. In: Proceedings of 2022 30th European Signal Processing Conference (EUSIPCO). 2022 30th European Signal Processing Conference (EUSIPCO), 29 Aug - 02 Sep 2022, Belgrade, Serbia. Institute of Electrical and Electronics Engineers (IEEE) , Belgrade, Serbia , pp. 165-169. ISBN 9781665467995
Abstract
Training of speech enhancement systems often does not incorporate knowledge of human perception and thus can lead to unnatural sounding results. Incorporating psychoacoustically motivated speech perception metrics as part of model training via a predictor network has recently gained interest. However, the performance of such predictors is limited by the distribution of metric scores that appear in the training data. In this work, we propose MetricGAN+/- (an extension of Metric-GAN+, one such metric-motivated system) which introduces an additional network - a “de-generator” to improve the robustness of the prediction network (and by extension of the generator) by ensuring observation of a wider range of metric scores in training. Experimental results on the VoiceBank-DEMAND dataset show relative improvement in PESQ score of 3.8% (3.05 vs. 3.22 PESQ score), as well as better generalisation to unseen noise and speech signals.
Metadata
Item Type: | Proceedings Paper |
---|---|
Authors/Creators: |
|
Copyright, Publisher and Additional Information: | © 2022 by European Association for Signal Processing (EURASIP). |
Keywords: | speech enhancement; noise reduction; speech quality metrics; neural networks; GAN; metric prediction |
Dates: |
|
Institution: | The University of Sheffield |
Academic Units: | The University of Sheffield > Faculty of Engineering (Sheffield) > Department of Computer Science (Sheffield) |
Funding Information: | Funder Grant number Engineering and Physical Sciences Research Council 2429310 |
Depositing User: | Symplectic Sheffield |
Date Deposited: | 15 Nov 2022 11:49 |
Last Modified: | 15 Nov 2022 11:52 |
Published Version: | https://ieeexplore.ieee.org/document/9909682 |
Status: | Published |
Publisher: | Institute of Electrical and Electronics Engineers (IEEE) |
Refereed: | Yes |
Related URLs: | |
Open Archives Initiative ID (OAI ID): | oai:eprints.whiterose.ac.uk:193416 |
Available Versions of this Item
-
MetricGAN+/- : increasing robustness of noise reduction on unseen data. (deposited 22 Jun 2022 09:46)
- MetricGAN+/-: increasing robustness of noise reduction on unseen data. (deposited 15 Nov 2022 11:49) [Currently Displayed]