用双向Mamba替代自注意力，实现实时语音深度伪造检测。
Fake-Mamba: Real-Time Speech Deepfake Detection Using Bidirectional Mamba as Self-Attention's Alternative
- 用双向Mamba替代自注意力，捕捉语音中的局部与全局异常
- 在三个基准上分别达0.97%、1.74%、5.85%错误率，优于现有模型
- 支持实时推理，适合部署于实际安防场景
语音合成技术的进步加剧了安全威胁，推动了实时深度伪造检测研究。本文探讨双向Mamba能否作为自注意力的有效替代方案来识别合成语音。提出的Fake-Mamba框架结合XLSR前端与双向Mamba，以捕捉局部和全局异常特征。核心创新在于设计三种高效编码器：TransBiMamba、ConBiMamba和PN-BiMamba。利用XLSR丰富的语言表征，PN-BiMamba能有效捕获合成语音的细微线索。在ASVspoof 21 LA、21 DF和In-The-Wild基准上，Fake-Mamba分别达到0.97%、1.74%和5.85%的等错误率（EER），相比SOTA模型XLSR-Conformer和XLSR-Mamba有显著提升。该框架在不同语句长度下保持实时推理能力，展现出强泛化性和实际应用潜力。代码已公开于https://github.com/xuanxixi/Fake-Mamba。
原文摘要 · Abstract (English)
Advances in speech synthesis intensify security threats, motivating real-time deepfake detection research. We investigate whether bidirectional Mamba can serve as a competitive alternative to Self-Attention in detecting synthetic speech. Our solution, Fake-Mamba, integrates an XLSR front-end with bidirectional Mamba to capture both local and global artifacts. Our core innovation introduces three efficient encoders: TransBiMamba, ConBiMamba, and PN-BiMamba. Leveraging XLSR's rich linguistic representations, PN-BiMamba can effectively capture the subtle cues of synthetic speech. Evaluated on ASVspoof 21 LA, 21 DF, and In-The-Wild benchmarks, Fake-Mamba achieves 0.97%, 1.74%, and 5.85% EER, respectively, representing substantial relative gains over SOTA models XLSR-Conformer and XLSR-Mamba. The framework maintains real-time inference across utterance lengths, demonstrating strong generalization and practical viability. The code is available at https://github.com/xuanxixi/Fake-Mamba.
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。