dc.contributorDhirubhai Ambani Institute of Information and Communication Technology (DA-IICT)
dc.contributorUniversidade Estadual Paulista (UNESP)
dc.date.accessioned2022-05-01T09:00:56Z
dc.date.accessioned2022-12-20T03:41:39Z
dc.date.available2022-05-01T09:00:56Z
dc.date.available2022-12-20T03:41:39Z
dc.date.created2022-05-01T09:00:56Z
dc.date.issued2022-03-01
dc.identifierComputer Speech and Language, v. 72.
dc.identifier1095-8363
dc.identifier0885-2308
dc.identifierhttp://hdl.handle.net/11449/233527
dc.identifier10.1016/j.csl.2021.101281
dc.identifier2-s2.0-85114778313
dc.identifier.urihttps://repositorioslatinoamericanos.uchile.cl/handle/2250/5413626
dc.description.abstractIn 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.
dc.languageeng
dc.relationComputer Speech and Language
dc.sourceScopus
dc.subjectAutomatic speaker verification (ASV)
dc.subjectEnhanced Teager Energy Cepstral Coefficients (ETECCs)
dc.subjectEnhanced Teager Energy Operator (ETEO)
dc.subjectHandcrafted features
dc.subjectParaconsistent Feature Engineering (PFE)
dc.subjectReplay attacks (RAs)
dc.titleImproving the potential of Enhanced Teager Energy Cepstral Coefficients (ETECC) for replay attack detection
dc.typeArtículos de revistas


Este ítem pertenece a la siguiente institución