mlsanityData to 19 Oct 2026

Joint Optimization of ASV and CM tasks: BTUEF Team's Submission for WildSpoof Challenge

arXiv 2602.01722 ↗First posted 2 Feb 2026eess.AScited by 0

On the map: Speech and audio language models → Anti-spoofing and adversarial audio attacks

Abstract (arXiv)

Spoofing-aware speaker verification (SASV) jointly addresses automatic speaker verification and spoofing countermeasures to improve robustness against adversarial attacks. In this paper, we investigate our recently proposed modular SASV framework that enables effective reuse of publicly available ASV and CM systems through non-linear fusion, explicitly modeling their interaction, and optimization with an operating-condition-dependent trainable a-DCF loss. The framework is evaluated using ECAPA-TDNN and ReDimNet as ASV embedding extractors and SSL-AASIST as the CM model, with experiments conducted both with and without fine-tuning on the WildSpoof SASV training data. Results show that the best performance is achieved by combining ReDimNet-based ASV embeddings with fine-tuned SSL-AASIST representations, yielding an a-DCF of 0.0515 on the progress evaluation set and 0.2163 on the final evaluation set.

Key facts. No citing papers in the map yet (data through 19 Oct 2026). Placed in Anti-spoofing and adversarial audio attacks, one of 11 directions in Speech and audio language models. Builds on 10 of its 12 references that are in the map. First posted on arXiv 2 Feb 2026.

Builds on

10 of 12 references.

Closest ancestors · 2

Same direction, newest first

  1. 2025Joint Optimization of Speaker and Spoof Detectors for Spoofing-Robust Automatic Speaker Verification2 Oct
  2. 2024Spoofing-Robust Speaker Verification Using Parallel Embedding Fusion: BTU Speech Group's Approach for ASVspoof5 Challenge28 Aug

Other references · 8

  1. 2024ASVspoof 5: Crowdsourced Speech Data, Deepfakes, and Adversarial Attacks at Scalecited by 121
  2. 2020The Attacker's Perspective on Automatic Speaker Verification: An Overviewcited by 33
  3. 2022SASV 2022: The First Spoofing-Aware Speaker Verification Challengecited by 31
  4. 2024a-DCF: an architecture agnostic metric with application to spoofing-robust speaker verificationcited by 17
  5. 2024Revisiting and Improving Scoring Fusion for Spoofing-aware Speaker Verification Using Compositional Data Analysiscited by 9
  6. 2025Replay Attacks Against Audio Deepfake Detectioncited by 8
  7. 2025WildSpoof Challenge Evaluation Plancited by 7
  8. 2024Optimizing a-DCF for Spoofing-Robust Speaker Verificationcited by 3

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2022Tackling Spoofing-Aware Speaker Verification with Multi-Model Fusion
  2. 2022Baseline Systems for the First Spoofing-Aware Speaker Verification Challenge: Score and Embedding Fusion
  3. 2025BUT Systems for WildSpoof Challenge: SASV in the Wild
  4. 2022Spoofing-Aware Speaker Verification by Multi-Level Fusion
  5. 2024Generalizing Speaker Verification for Spoof Awareness in the Embedding Space
  6. 2023Towards single integrated spoofing-aware speaker verification embeddings
  7. 2022A Probabilistic Fusion Framework for Spoofing Aware Speaker Verification
  8. 2022Spoofing-Aware Attention based ASV Back-end with Multiple Enrollment Utterances and a Sampling Strategy for the SASV Challenge 2022