用频谱重缩放和Wav2Vec2攻击语音匿名系统，暴露其安全漏洞。

SpecWav-Attack: Leveraging Spectrogram Resizing and Wav2Vec 2.0 for Attacking Anonymized Speech

• 结合Wav2Vec2与频谱重缩放提取特征，提升攻击精度
• 在LibriSpeech数据集上超越传统攻击方法，验证有效性
• 适合研究语音隐私与对抗攻击的学者参考

本文提出SpecWav-Attack，一种用于检测匿名语音中说话人身份的对抗模型。该方法利用Wav2Vec2进行特征提取，并引入频谱重缩放与增量训练策略以提升性能。在LibriSpeech-dev和LibriSpeech-test数据集上的评估显示，该方法优于传统攻击手段，揭示了现有语音匿名系统存在的安全隐患，强调需加强防护措施，并以ICASSP 2025攻击挑战赛为基准进行对比。

原文摘要 · Abstract (English)

This paper presents SpecWav-Attack, an adversarial model for detecting speakers in anonymized speech. It leverages Wav2Vec2 for feature extraction and incorporates spectrogram resizing and incremental training for improved performance. Evaluated on librispeech-dev and librispeech-test, SpecWav-Attack outperforms conventional attacks, revealing vulnerabilities in anonymized speech systems and emphasizing the need for stronger defenses, benchmarked against the ICASSP 2025 Attacker Challenge.

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。