不依赖标注，通过分析语音帧间时序差异检测部分伪造语音。
Frame-level Temporal Difference Learning for Partial Deepfake Speech Detection
- 从帧级时序差异出发，捕捉伪造语音的异常变化模式。
- 在PartialSpoof和HAD数据集上误报率低至0.59%和0.03%。
- 适合无标注数据场景下的深度伪造语音检测应用。
由于潜在的隐蔽误导风险，检测部分深度伪造语音至关重要。然而，现有方法依赖昂贵的帧级标注进行训练，限制了实际可扩展性。同时，它们主要关注真实与伪造片段之间的过渡伪影，但随着生成技术使过渡更加平滑，检测难度上升。本文提出新视角：分析帧级时序差异，发现伪造语音相比真实语音具有更剧烈的方向性波动和不自然的局部变化。基于此，我们设计了时序差分注意力模块（TDAM），将部分深度伪造检测转化为识别异常时序变化，无需显式边界标注。该模块采用双层级层次化差异表示，在细粒度与粗粒度尺度上捕捉时序不规则性；自适应平均池化则保留可变长度输入中的关键模式，减少信息损失。所提TDAM-AvgPool模型在PartialSpoof数据集上达到0.59%的EER，HAD数据集上为0.03%，显著优于现有方法，且无需帧级监督。
原文摘要 · Abstract (English)
Detecting partial deepfake speech is essential due to its potential for subtle misinformation. However, existing methods depend on costly frame-level annotations during training, limiting real-world scalability. Also, they focus on detecting transition artifacts between bonafide and deepfake segments. As deepfake generation techniques increasingly smooth these transitions, detection has become more challenging. To address this, our work introduces a new perspective by analyzing frame-level temporal differences and reveals that deepfake speech exhibits erratic directional changes and unnatural local transitions compared to bonafide speech. Based on this finding, we propose a Temporal Difference Attention Module (TDAM) that redefines partial deepfake detection as identifying unnatural temporal variations, without relying on explicit boundary annotations. A dual-level hierarchical difference representation captures temporal irregularities at both fine and coarse scales, while adaptive average pooling preserves essential patterns across variable-length inputs to minimize information loss. Our TDAM-AvgPool model achieves state-of-the-art performance, with an EER of 0.59% on the PartialSpoof dataset and 0.03% on the HAD dataset, which significantly outperforms the existing methods without requiring frame-level supervision.
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。