Optimizing a-DCF for Spoofing-Robust Speaker Verification

Basit Öğe Kaydı
dc.authorid0000-0002-9174-0367
dc.authorid0000-0002-4371-7322
dc.contributor.authorKurnaz, Oguzhan
dc.contributor.authorMishra, Jagabandhu
dc.contributor.authorKinnunen, Tomi H.
dc.contributor.authorHanilci, Cemal
dc.date.accessioned2026-02-08T15:15:41Z
dc.date.available2026-02-08T15:15:41Z
dc.date.issued2025
dc.departmentBursa Teknik Üniversitesi
dc.description.abstractAutomatic speaker verification (ASV) systems are vulnerable to spoofing attacks. We propose a spoofing-robust ASV system optimized directly for the recently introduced architecture-agnostic detection cost function (a-DCF), which allows targeting a desired trade-off between the contradicting aims of user convenience and robustness to spoofing. We combine a-DCF and binary cross-entropy (BCE) with a novel straightforward threshold optimization technique. Our results with an embedding fusion system on ASVspoof2019 data demonstrate relative improvement of 13% over a system trained using BCE only (from minimum a-DCF of 0.1445 to 0.1254). Using an alternative non-linear score fusion approach provides relative improvement of 43% (from minimum a-DCF of 0.0508 to 0.0289).
dc.description.sponsorshipAcademy of Finland [349605]; Scientific and Technological Research Council of Turkiye (TUBITAK) [1059B142300330]
dc.description.sponsorshipThis work was supported by the Academy of Finland under Grant 349605, through Project SPEECHFAKES. The work of Oguzhan Kurnaz was supported by the Scientific and Technological Research Council of Turkiye (TUBITAK) through the 2214-A fellowship program under Grant 1059B142300330.
dc.identifier.doi10.1109/LSP.2025.3545290
dc.identifier.endpage1085
dc.identifier.issn1070-9908
dc.identifier.issn1558-2361
dc.identifier.scopus2-s2.0-105001060545
dc.identifier.scopusqualityQ1
dc.identifier.startpage1081
dc.identifier.urihttps://doi.org/10.1109/LSP.2025.3545290
dc.identifier.urihttps://hdl.handle.net/20.500.12885/5905
dc.identifier.volume32
dc.identifier.wosWOS:001445057900002
dc.identifier.wosqualityQ2
dc.indekslendigikaynakWeb of Science
dc.indekslendigikaynakScopus
dc.language.isoen
dc.publisherIeee-Inst Electrical Electronics Engineers Inc
dc.relation.ispartofIeee Signal Processing Letters
dc.relation.publicationcategoryMakale - Uluslararası Hakemli Dergi - Kurum Öğretim Elemanı
dc.rightsinfo:eu-repo/semantics/closedAccess
dc.snmzWOS_KA_20260207
dc.subjectMeasurement
dc.subjectCosts
dc.subjectTraining
dc.subjectCost function
dc.subjectSignal processing algorithms
dc.subjectError analysis
dc.subjectSecurity
dc.subjectRobustness
dc.subjectComputer architecture
dc.subjectTraining data
dc.subjecta-DCF
dc.subjectspoofing-robust speaker verification
dc.titleOptimizing a-DCF for Spoofing-Robust Speaker Verification
dc.typeArticle

Dosyalar

Koleksiyon

WoS İndeksli Yayınlar Koleksiyonu
Scopus İndeksli Yayınlar Koleksiyonu