ACES：用于自动语音识别中耦合、解释和压力测试的声调子空间
ACES: Accent Subspaces for Coupling, Explanations, and Stress-Testing in Automatic Speech Recognition
AI总结：
ACES通过提取口音区分子空间，揭示ASR中口音差异的深层结构，证明其对公平性评估的重要性。
摘要：
ASR系统在不同口音之间表现出持续的性能差异，但这些差距是表面偏见还是深层结构漏洞仍不清楚。我们引入ACES，一种三阶段审计方法，从ASR表示中提取口音区分子空间，将对抗攻击限制在其中，并测试移除这些子空间是否能提高公平性。在Wav2Vec2-base模型上，不可察觉的扰动（约60 dB SNR）沿口音子空间使WER差异差距增加近50%（21.3→31.8 pp），超过随机子空间对照组；一个排列标签测试证实其对真实口音结构的特异性。部分移除子空间会同时恶化WER和差异，揭示出口音区分和识别关键特征深度交织。因此，ACES将口音子空间定位为强大的公平性审计工具，而非简单的删除杠杆。
英文摘要：
ASR systems exhibit persistent performance disparities across accents, but whether these gaps reflect superficial biases or deep structural vulnerabilities remains unclear. We introduce ACES, a three-stage audit that extracts accent-discriminative subspaces from ASR representations, constrains adversarial attacks to them, and tests whether removing them improves fairness. On Wav2Vec2-base with seven accents, imperceptible perturbations (~60 dB SNR) along the accent subspace amplify the WER disparity gap by nearly 50% (21.3->31.8 pp), exceeding random-subspace controls; a permuted-label test confirms specificity to genuine accent structure. Partially removing the subspace worsens both WER and disparity, revealing that accent-discriminative and recognition-critical features are deeply entangled. ACES thus positions accent subspaces as powerful fairness-auditing tools, not simple erasure levers.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
