Improving the potential of Enhanced Teager Energy Cepstral Coefficients (ETECC) for replay attack detection
Autor(a) principal: | |
---|---|
Data de Publicação: | 2022 |
Outros Autores: | , , |
Tipo de documento: | Artigo |
Idioma: | eng |
Título da fonte: | Repositório Institucional da UNESP |
Texto Completo: | http://dx.doi.org/10.1016/j.csl.2021.101281 http://hdl.handle.net/11449/233527 |
Resumo: | In the scope of voice biometrics, the term replay attack, (RA) refers to the dishonest attempt made by an impostor to spoof someone else's identity by replaying the subject's previously recorded speech close to the Automatic Speaker Verification (ASV) system under attack. State-of-the-art strategies for RA detection, such as the Enhanced Teager Energy Cepstral Coefficients (ETECC), have shown promising results due to their precision in measuring energy from high frequency components of speech, as a function of two recently defined concepts: signal mass and Enhanced Teager Energy Operator (ETEO). Nevertheless, since the replay mechanism prominently deteriorates the speech signal spectrum just in those spectral zones, we propose the association of ETEO with different strategies to further improve the previous results in getting effective countermeasures against RAs. Specifically, comprehensive evaluations which include a detailed mathematical analysis, a simulation on amplitude and frequency modulated (AM–FM) signals, and a spectrographic inspection involving different filterbank structures, along with their experimental results, are provided in this paper. In addition, ETEO-derived features are contrasted to existing feature sets by using Paraconsistent Feature Engineering (PFE) for feature ranking, expanding our previously published results. Lastly, experiments are performed with ASVSpoof-2017 version 2.0 dataset, Realistic Replay Attack Microphone Array Speech Corpus (ReMASC), BTAS-2016, dataset, ASVSpoof-2019 challenge dataset, and ASVSpoof-2015 challenge dataset, considering Gaussian Mixture Models (GMMs), Convolutional Neural Networks (CNNs) and Light-CNN architectures as being the classifiers. The standalone ETECC-GMM system showed the best performance by producing equal error rates (EERs) of 5.55% and 10.75% on development and evaluation sets, respectively. |
id |
UNSP_051de06c71b8b5e3dbba27f87342e8f0 |
---|---|
oai_identifier_str |
oai:repositorio.unesp.br:11449/233527 |
network_acronym_str |
UNSP |
network_name_str |
Repositório Institucional da UNESP |
repository_id_str |
2946 |
spelling |
Improving the potential of Enhanced Teager Energy Cepstral Coefficients (ETECC) for replay attack detectionAutomatic speaker verification (ASV)Enhanced Teager Energy Cepstral Coefficients (ETECCs)Enhanced Teager Energy Operator (ETEO)Handcrafted featuresParaconsistent Feature Engineering (PFE)Replay attacks (RAs)In the scope of voice biometrics, the term replay attack, (RA) refers to the dishonest attempt made by an impostor to spoof someone else's identity by replaying the subject's previously recorded speech close to the Automatic Speaker Verification (ASV) system under attack. State-of-the-art strategies for RA detection, such as the Enhanced Teager Energy Cepstral Coefficients (ETECC), have shown promising results due to their precision in measuring energy from high frequency components of speech, as a function of two recently defined concepts: signal mass and Enhanced Teager Energy Operator (ETEO). Nevertheless, since the replay mechanism prominently deteriorates the speech signal spectrum just in those spectral zones, we propose the association of ETEO with different strategies to further improve the previous results in getting effective countermeasures against RAs. Specifically, comprehensive evaluations which include a detailed mathematical analysis, a simulation on amplitude and frequency modulated (AM–FM) signals, and a spectrographic inspection involving different filterbank structures, along with their experimental results, are provided in this paper. In addition, ETEO-derived features are contrasted to existing feature sets by using Paraconsistent Feature Engineering (PFE) for feature ranking, expanding our previously published results. Lastly, experiments are performed with ASVSpoof-2017 version 2.0 dataset, Realistic Replay Attack Microphone Array Speech Corpus (ReMASC), BTAS-2016, dataset, ASVSpoof-2019 challenge dataset, and ASVSpoof-2015 challenge dataset, considering Gaussian Mixture Models (GMMs), Convolutional Neural Networks (CNNs) and Light-CNN architectures as being the classifiers. The standalone ETECC-GMM system showed the best performance by producing equal error rates (EERs) of 5.55% and 10.75% on development and evaluation sets, respectively.Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)Speech Research Lab Dhirubhai Ambani Institute of Information and Communication Technology (DA-IICT)Instituto de Biociências Letras e Ciências Exatas Unesp - Univ Estadual Paulista (São Paulo State University), Rua Cristóvão Colombo 2265, Jd NazarethInstituto de Biociências Letras e Ciências Exatas Unesp - Univ Estadual Paulista (São Paulo State University), Rua Cristóvão Colombo 2265, Jd NazarethFAPESP: 2019/04475-0FAPESP: 306808/2018-8Dhirubhai Ambani Institute of Information and Communication Technology (DA-IICT)Universidade Estadual Paulista (UNESP)Patil, Ankur T.Acharya, RajulPatil, Hemant A.Guido, Rodrigo Capobianco [UNESP]2022-05-01T09:00:56Z2022-05-01T09:00:56Z2022-03-01info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/articlehttp://dx.doi.org/10.1016/j.csl.2021.101281Computer Speech and Language, v. 72.1095-83630885-2308http://hdl.handle.net/11449/23352710.1016/j.csl.2021.1012812-s2.0-85114778313Scopusreponame:Repositório Institucional da UNESPinstname:Universidade Estadual Paulista (UNESP)instacron:UNESPengComputer Speech and Languageinfo:eu-repo/semantics/openAccess2022-05-01T09:00:56Zoai:repositorio.unesp.br:11449/233527Repositório InstitucionalPUBhttp://repositorio.unesp.br/oai/requestopendoar:29462024-08-05T15:58:38.802562Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP)false |
dc.title.none.fl_str_mv |
Improving the potential of Enhanced Teager Energy Cepstral Coefficients (ETECC) for replay attack detection |
title |
Improving the potential of Enhanced Teager Energy Cepstral Coefficients (ETECC) for replay attack detection |
spellingShingle |
Improving the potential of Enhanced Teager Energy Cepstral Coefficients (ETECC) for replay attack detection Patil, Ankur T. Automatic speaker verification (ASV) Enhanced Teager Energy Cepstral Coefficients (ETECCs) Enhanced Teager Energy Operator (ETEO) Handcrafted features Paraconsistent Feature Engineering (PFE) Replay attacks (RAs) |
title_short |
Improving the potential of Enhanced Teager Energy Cepstral Coefficients (ETECC) for replay attack detection |
title_full |
Improving the potential of Enhanced Teager Energy Cepstral Coefficients (ETECC) for replay attack detection |
title_fullStr |
Improving the potential of Enhanced Teager Energy Cepstral Coefficients (ETECC) for replay attack detection |
title_full_unstemmed |
Improving the potential of Enhanced Teager Energy Cepstral Coefficients (ETECC) for replay attack detection |
title_sort |
Improving the potential of Enhanced Teager Energy Cepstral Coefficients (ETECC) for replay attack detection |
author |
Patil, Ankur T. |
author_facet |
Patil, Ankur T. Acharya, Rajul Patil, Hemant A. Guido, Rodrigo Capobianco [UNESP] |
author_role |
author |
author2 |
Acharya, Rajul Patil, Hemant A. Guido, Rodrigo Capobianco [UNESP] |
author2_role |
author author author |
dc.contributor.none.fl_str_mv |
Dhirubhai Ambani Institute of Information and Communication Technology (DA-IICT) Universidade Estadual Paulista (UNESP) |
dc.contributor.author.fl_str_mv |
Patil, Ankur T. Acharya, Rajul Patil, Hemant A. Guido, Rodrigo Capobianco [UNESP] |
dc.subject.por.fl_str_mv |
Automatic speaker verification (ASV) Enhanced Teager Energy Cepstral Coefficients (ETECCs) Enhanced Teager Energy Operator (ETEO) Handcrafted features Paraconsistent Feature Engineering (PFE) Replay attacks (RAs) |
topic |
Automatic speaker verification (ASV) Enhanced Teager Energy Cepstral Coefficients (ETECCs) Enhanced Teager Energy Operator (ETEO) Handcrafted features Paraconsistent Feature Engineering (PFE) Replay attacks (RAs) |
description |
In the scope of voice biometrics, the term replay attack, (RA) refers to the dishonest attempt made by an impostor to spoof someone else's identity by replaying the subject's previously recorded speech close to the Automatic Speaker Verification (ASV) system under attack. State-of-the-art strategies for RA detection, such as the Enhanced Teager Energy Cepstral Coefficients (ETECC), have shown promising results due to their precision in measuring energy from high frequency components of speech, as a function of two recently defined concepts: signal mass and Enhanced Teager Energy Operator (ETEO). Nevertheless, since the replay mechanism prominently deteriorates the speech signal spectrum just in those spectral zones, we propose the association of ETEO with different strategies to further improve the previous results in getting effective countermeasures against RAs. Specifically, comprehensive evaluations which include a detailed mathematical analysis, a simulation on amplitude and frequency modulated (AM–FM) signals, and a spectrographic inspection involving different filterbank structures, along with their experimental results, are provided in this paper. In addition, ETEO-derived features are contrasted to existing feature sets by using Paraconsistent Feature Engineering (PFE) for feature ranking, expanding our previously published results. Lastly, experiments are performed with ASVSpoof-2017 version 2.0 dataset, Realistic Replay Attack Microphone Array Speech Corpus (ReMASC), BTAS-2016, dataset, ASVSpoof-2019 challenge dataset, and ASVSpoof-2015 challenge dataset, considering Gaussian Mixture Models (GMMs), Convolutional Neural Networks (CNNs) and Light-CNN architectures as being the classifiers. The standalone ETECC-GMM system showed the best performance by producing equal error rates (EERs) of 5.55% and 10.75% on development and evaluation sets, respectively. |
publishDate |
2022 |
dc.date.none.fl_str_mv |
2022-05-01T09:00:56Z 2022-05-01T09:00:56Z 2022-03-01 |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/article |
format |
article |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
http://dx.doi.org/10.1016/j.csl.2021.101281 Computer Speech and Language, v. 72. 1095-8363 0885-2308 http://hdl.handle.net/11449/233527 10.1016/j.csl.2021.101281 2-s2.0-85114778313 |
url |
http://dx.doi.org/10.1016/j.csl.2021.101281 http://hdl.handle.net/11449/233527 |
identifier_str_mv |
Computer Speech and Language, v. 72. 1095-8363 0885-2308 10.1016/j.csl.2021.101281 2-s2.0-85114778313 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.relation.none.fl_str_mv |
Computer Speech and Language |
dc.rights.driver.fl_str_mv |
info:eu-repo/semantics/openAccess |
eu_rights_str_mv |
openAccess |
dc.source.none.fl_str_mv |
Scopus reponame:Repositório Institucional da UNESP instname:Universidade Estadual Paulista (UNESP) instacron:UNESP |
instname_str |
Universidade Estadual Paulista (UNESP) |
instacron_str |
UNESP |
institution |
UNESP |
reponame_str |
Repositório Institucional da UNESP |
collection |
Repositório Institucional da UNESP |
repository.name.fl_str_mv |
Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP) |
repository.mail.fl_str_mv |
|
_version_ |
1808128588955779072 |