首个多语言伪造音频仇恨言论检测数据集，助力跨语言安全防护。
SynHate: Detecting Hate Speech in Synthetic Deepfake Audio
- 构建四类合成音频数据集，覆盖37种语言，涵盖真实与伪造的正常及仇恨内容。
- 测试五种自监督模型发现，Whisper-small在多数语言上表现最佳，跨数据集泛化仍弱。
- 适合研究多语言仇恨言论检测、语音安全与文化敏感性模型的开发者使用。
深度伪造音频与仇恨言论的结合，借助先进文本转语音技术，严重威胁网络环境安全。我们提出SynHate，首个用于检测合成音频中仇恨言论的多语言数据集，覆盖37种语言。该数据集采用创新的四分类体系：真实正常、真实仇恨、伪造正常、伪造仇恨。基于MuTox和ADIMA数据集构建，涵盖全球及印度地区的多样化仇恨言论模式。我们评估了五种主流自监督模型（Whisper-small/medium、XLS-R、AST、mHuBERT），发现不同语言间性能差异显著，Whisper-small整体表现最优。跨数据集泛化能力仍面临挑战。通过发布SynHate数据集与基线代码，我们旨在推动更鲁棒、具文化敏感性且多语言的合成仇恨言论应对方案。数据集可于https://www.iab-rubric.org/resources获取。
原文摘要 · Abstract (English)
The rise of deepfake audio and hate speech, powered by advanced text-to-speech, threatens online safety. We present SynHate, the first multilingual dataset for detecting hate speech in synthetic audio, spanning 37 languages. SynHate uses a novel four-class scheme: Real-normal, Real-hate, Fake-normal, and Fake-hate. Built from MuTox and ADIMA datasets, it captures diverse hate speech patterns globally and in India. We evaluate five leading self-supervised models (Whisper-small/medium, XLS-R, AST, mHuBERT), finding notable performance differences by language, with Whisper-small performing best overall. Cross-dataset generalization remains a challenge. By releasing SynHate and baseline code, we aim to advance robust, culturally sensitive, and multilingual solutions against synthetic hate speech. The dataset is available at https://www.iab-rubric.org/resources.
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。