在浏览器插件中实现轻量级音频深度伪造检测，兼顾速度与隐私。
Detecting Audio Deepfakes on the Edge:Lightweight SSL-Based Detection in a Browser Plugin
- 用截断的自监督模型加逻辑分类器，实现快速检测。
- 相比基线AASIST提升10%准确率，推理速度加快40%。
- 本地运行不依赖云端，适合记者和核查人员使用。
音频深度伪造正对公众、记者及事实核查人员构成日益严峻的挑战。后者需要可靠的工具来验证信息来源的真实性，同时保护信息隐私。现有商业检测方案依赖云处理，引发隐私担忧。为此，我们提出一种在设备端运行的音频深度伪造检测模型。结果表明，采用截断的自监督骨干网络与简单逻辑分类器，不仅运行极快，且准确率常优于现有方案。该模型相较基线AASIST提升10%准确率，推理速度加快40%。我们将其集成至浏览器插件中，使记者与核查人员可安全便捷地检测深度伪造音频。插件代码已开源：https://github.com/OctavianPascu97/Audio-Deepfakes-Browser-Plugin。
原文摘要 · Abstract (English)
Audio deepfakes are a growing challenge for the general public, as well as for journalists and fact-checkers. The latter need reliable tools to verify the authenticity of their sources, while at the same time keeping their information private. Commercial deepfake detection solutions rely on cloud-based processing, which raises privacy concerns. To solve this problem, we propose an on-device audio deepfake detection model. We show that a truncated self-supervised backbone with a simple logistic classifier is both very fast and often more accurate than existing solutions. Our solution outperforms the baseline AASIST by 10% and improves inference speed by 40%. We integrate this model into a browser plug-in, which allows journalists and fact-checkers to detect deepfakes easily and securely. Code for the plugin is available at https://github.com/OctavianPascu97/Audio-Deepfakes-Browser-Plugin.
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。