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

arXiv:2609.16458v1 Announce Type: cross 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.

Few-Shot Contrastive Adaptation for Audio Abuse Detection in Low-Resource Indic Languages
Motor, Cognitive, or Corpus? What Survives Cross-Lingual Transfer in Speech-Based Parkinsons Disease Detection
Cross-lingual Matryoshka Representation Learning across Speech and Text
Multilingual Detection of Alzheimer's Disease from Speech: A Cross-Linguistic Transfer Learning Approach
Disentangling Speaker and Language Effects in Cross-Lingual Speaker Verification for Iberian Languages

Source: arXiv cs.CL | 2026-09-16

Related