Joint Optimization Of Speaker And Spoof Detectors For Spoofing-robust Automatic Speaker Verification

Abstract

Spoofing-robust speaker verification (SASV) combines the tasks of speaker and spoof detection to authenticate speakers under adversarial settings. Many SASV systems rely on fusion of speaker and spoof cues at embedding, score or decision levels, based on independently trained subsystems. In this study, we respect similar modularity of the two subsystems, by integrating their outputs using trainable back-end classifiers. In particular, we explore various approaches for directly optimizing the back-end for the recently-proposed SASV performance metric (a-DCF) as a training objective. Our experiments on the ASVspoof 5 dataset demonstrate two important findings: (i) nonlinear score fusion consistently improves a-DCF over linear fusion, and (ii) the combination of weighted cosine scoring for speaker detection with SSL-AASIST for spoof detection achieves state-of-the-art performance, reducing min a-DCF to 0.196 and SPF-EER to 7.6%. These contributions highlight the importance of modular desi

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keykurnaz2025joint

Related papers

• Baseline Systems For The First Spoofing-aware Speaker Verification Challenge: Score And Embedding Fusion (2022)6.77
• Two Methods For Spoofing-aware Speaker Verification: Multi-layer Perceptron Score Fusion Model And Integrated Embedding Projector (2022)2.26
• Spoofing-robust Speaker Verification Using Parallel Embedding Fusion: BTU Speech Group's Approach For Asvspoof5 Challenge (2024)5.24
• SA-SASV: An End-to-end Spoof-aggregated Spoofing-aware Speaker Verification System (2022)6.77
• Optimizing A-dcf For Spoofing-robust Speaker Verification (2024)3.58
• Spoofing-aware Attention Based ASV Back-end With Multiple Enrollment Utterances And A Sampling Strategy For The SASV Challenge 2022 (2022)4.52
• Anti-spoofing Methods For Automatic Speakerverification System (2017)2.26
• Generalizing Speaker Verification For Spoof Awareness In The Embedding Space (2024)7.16