研究人员开发了新的基准来解决语音欺骗检测系统中的泛化差距问题，这些系统难以跟上先进的 LLM 驱动的文本到语音和语音转换技术。VoxENES 2026 基准包含 10 种当代语音合成方法生成的超过 53,000 个英语和西班牙语音频样本，揭示了现有检测器性能的显著下降。同样，PC-Mix 数据集解决了在混合语音和环境声音中检测欺骗音频组件的挑战，而这些条件在先前的研究中常常被忽视。 AI
影响 强调需要更强大的 AI 驱动的音频欺骗检测方法来对抗先进的合成语音。
排序理由 两篇研究论文介绍了用于音频欺骗检测的新基准和数据集。
AI 生成摘要 · Google Gemini · 来自 3 个来源。 我们如何撰写摘要 →
arXiv:2607.11706v1 Announce Type: cross Abstract: Modern LLM-driven text-to-speech (TTS) and voice conversion (VC) systems produce synthetic speech that differs from the generators represented in many legacy spoofing benchmarks. This mismatch creates a temporal generalization gap…
arXiv:2607.10345v1 Announce Type: cross Abstract: Recent studies on partial audio spoofing mainly focus on studio-recorded speech with temporal localization of spoofed segments. However, these studies often overlook realistic conditions where spoofed and bonafide segments simulta…
Modern LLM-driven text-to-speech (TTS) and voice conversion (VC) systems produce synthetic speech that differs from the generators represented in many legacy spoofing benchmarks. This mismatch creates a temporal generalization gap that can overestimate detector robustness under r…