Language Orthogonalization for Zero-Shot Cross-Lingual Audio Deepfake Detection

2026-09-14

AI Summary

本文通过语言正交化方法解决跨语言音频深度伪造检测中的零样本迁移问题，减少了自监督语音模型中的语言依赖性结构干扰。

Abstract

Audio deepfake detectors need to transfer to languages absent from training, as multilingual speech synthesis outpaces labeled anti-spoofing resources. While detectors increasingly rely on self-supervised speech models (S3Ms), these backbones encode language-dependent structure that confounds spoof cues. We address this confound through language orthogonalization, a target-free ridge map that removes S3M variation projected onto continuous language-identification (LID) embeddings. Across six languages, six S3M backbones, and all Leave-N-Out settings, it consistently reduces EER across unseen languages. Cross-lingual EER correlates with LID-space distance, where orthogonalization yields larger gains for more distant transfers.

Problem

• Audio Deepfake Detection
• Cross-Lingual
• Language Orthogonalization
• Self-Supervised Speech Models

Innovation

• Language Orthogonalization
• Self-supervised Speech Models
• Cross-lingual Audio Deepfake Detection
• Language-identification Embeddings