用评分调控嵌入，提升语音伪造检测的说话人验证鲁棒性。
ELEAT-SAGA: Early & Late Integration with Evading Alternating Training for Spoof-Robust Speaker Verification
- 基于反伪造评分动态调节说话人特征，实现智能注意力融合。
- 在ASVspoof 2019上达到1.22%的SASV-EER和0.0304的min a-DCF。
- 适合关注语音安全与对抗攻击的系统设计者阅读。
防伪造自动说话人验证（SASV）旨在构建对零努力欺骗攻击及复杂伪造技术（如语音转换和文本转语音）具备鲁棒性的系统。本文提出一种新型SASV架构SASV-SAGA，引入评分感知门控注意力（SAGA），根据反伪造（CM）评分动态调节说话人嵌入。通过整合预训练的ECAPA-TDNN和AASIST模型输出的说话人嵌入与CM评分，探索了早期、晚期及全融合等多种集成策略。进一步提出多模块交替训练（ATMM）及其改进版本——规避交替训练（EAT）。在ASVspoof 2019逻辑访问（LA）和Spoofceleb数据集上的实验表明，相比基线有显著提升，在ASVspoof 2019评测集上达到1.22%的欺骗感知说话人验证等错误率（SASV-EER）和0.0304的最小归一化无差别检测代价函数（min a-DCF），验证了评分感知注意力机制与交替训练策略在增强SASV系统鲁棒性方面的有效性。
原文摘要 · Abstract (English)
Spoofing-robust automatic speaker verification (SASV) seeks to build automatic speaker verification systems that are robust against both zero-effort impostor attacks and sophisticated spoofing techniques such as voice conversion (VC) and text-to-speech (TTS). In this work, we propose a novel SASV architecture that introduces score-aware gated attention (SAGA), SASV-SAGA, enabling dynamic modulation of speaker embeddings based on countermeasure (CM) scores. By integrating speaker embeddings and CM scores from pre-trained ECAPA-TDNN and AASIST models respectively, we explore several integration strategies including early, late, and full integration. We further introduce alternating training for multi-module (ATMM) and a refined variant, evading alternating training (EAT). Experimental results on the ASVspoof 2019 Logical Access (LA) and Spoofceleb datasets demonstrate significant improvements over baselines, achieving a spoofing aware speaker verification equal error rate (SASV-EER) of 1.22% and minimum normalized agnostic detection cost function (min a-DCF) of 0.0304 on the ASVspoof 2019 evaluation set. These results confirm the effectiveness of score-aware attention mechanisms and alternating training strategies in enhancing the robustness of SASV systems.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。