Skip to main contentSkip to search and navigation

UEF eREPOSITORY

    • English
    • suomi
  • English 
    • English
    • suomi
  • Login
View Item 
  •   Home
  • Artikkelit
  • Luonnontieteiden ja metsätieteiden tiedekunta
  • View Item
  •   Home
  • Artikkelit
  • Luonnontieteiden ja metsätieteiden tiedekunta
  • View Item
JavaScript is disabled for your browser. Some features of this site may not work without it.

Deep generative variational autoencoding for replay spoof detection in automatic speaker verification

Thumbnail
Files
Item embargoed until 2022-03-19. Restrictions imposed by the publisher
Self archived version
final draft
Date
2020
Author(s)
Chettri, Bhusan
Kinnunen, Tomi
Benetos, Emmanouil
Unique identifier
10.1016/j.csl.2020.101092
Metadata
Show full item record
More information
Research Database SoleCris

Self-archived article

Citation
Chettri, Bhusan. Kinnunen, Tomi. Benetos, Emmanouil. (2020). Deep generative variational autoencoding for replay spoof detection in automatic speaker verification.  Computer speech and language, 63, 101092. 10.1016/j.csl.2020.101092.
Rights
© 2020 Elsevier Ltd.
Licensed under
CC BY-NC-ND https://creativecommons.org/licenses/by-nc-nd/4.0/
Abstract

Automatic speaker verification (ASV) systems are highly vulnerable to presentation attacks, also called spoofing attacks. Replay is among the simplest attacks to mount — yet difficult to detect reliably. The generalization failure of spoofing countermeasures (CMs) has driven the community to study various alternative deep learning CMs. The majority of them are supervised approaches that learn a human-spoof discriminator. In this paper, we advocate a different, deep generative approach that leverages from powerful unsupervised manifold learning in classification. The potential benefits include the possibility to sample new data, and to obtain insights to the latent features of genuine and spoofed speech. To this end, we propose to use variational autoencoders (VAEs) as an alternative backend for replay attack detection, via three alternative models that differ in their class-conditioning. The first one, similar to the use of Gaussian mixture models (GMMs) in spoof detection, is to train independently two VAEs — one for each class. The second one is to train a single conditional model (C-VAE) by injecting a one-hot class label vector to the encoder and decoder networks. Our final proposal integrates an auxiliary classifier to guide the learning of the latent space. Our experimental results using constant-Q cepstral coefficient (CQCC) features on the ASVspoof 2017 and 2019 physical access subtask datasets indicate that the C-VAE offers substantial improvement in comparison to training two separate VAEs for each class. On the 2019 dataset, the C-VAE outperforms the VAE and the baseline GMM by an absolute 9 - 10% in both equal error rate (EER) and tandem detection cost function (t-DCF) metrics. Finally, we propose VAE residuals — the absolute difference of the original input and the reconstruction as features for spoofing detection. The proposed frontend approach augmented with a convolutional neural network classifier demonstrated substantial improvement over the VAE backend use case.

Subjects
anti-spoofing   presentation attack detection   replay attack   countermeasures   deep generative models   
URI
https://erepo.uef.fi/handle/123456789/8167
Link to the original item
http://dx.doi.org/10.1016/j.csl.2020.101092
Publisher
Elsevier BV
Collections
  • Luonnontieteiden ja metsätieteiden tiedekunta [1109]
University of Eastern Finland
OpenAccess
eRepo
erepo@uef.fi
OpenUEF
Service provided by
the University of Eastern Finland Library
Library web pages
Twitter
Facebook
Youtube
Library blog
 sitemap
Search

Browse

All of the ArchiveResource types & CollectionsBy Issue DateAuthorsTitlesSubjectsFacultyDepartmentFull organizationSeriesMain subjectThis CollectionBy Issue DateAuthorsTitlesSubjectsFacultyDepartmentFull organizationSeriesMain subject

My Account

Login
University of Eastern Finland
OpenAccess
eRepo
erepo@uef.fi
OpenUEF
Service provided by
the University of Eastern Finland Library
Library web pages
Twitter
Facebook
Youtube
Library blog
 sitemap