mlsanityData to 14 Oct 2026

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

arXiv 2609.16458 ↗First posted 15 Sep 2026eess.AS cs.CLcited by 1

Not on the map. Why?

Abstract (arXiv)

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.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 22 of its 25 references that are in the map. First posted on arXiv 15 Sep 2026.

Builds on

22 of 25 references.

References · 22

  1. 2020wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representationscited by 3,905
  2. 2021HuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden Unitscited by 2,342
  3. 2021WavLM: Large-Scale Self-Supervised Pre-Training for Full Stack Speech Processingcited by 1,657
  4. 2021XLS-R: Self-supervised Cross-lingual Speech Representation Learning at Scalecited by 601
  5. 2023Scaling Speech Technology to 1,000+ Languagescited by 411
  6. 2021Layer-wise Analysis of a Self-supervised Speech Representation Modelcited by 298
  7. 2019ASVspoof 2019: Future Horizons in Spoofed and Fake Audio Detectioncited by 178
  8. 2022Comparative layer-wise analysis of self-supervised speech modelscited by 158
  9. 2020VoxLingua107: a Dataset for Spoken Language Recognitioncited by 122
  10. 2022ASVspoof 2021: Towards Spoofed and Deepfake Speech Detection in the Wildcited by 121
  11. 2022Automatic speaker verification spoofing and deepfake detection using wav2vec 2.0 and data augmentationcited by 117
  12. 2024mHuBERT-147: A Compact Multilingual HuBERT Modelcited by 66
10 more
  1. 2024MLAAD: The Multi-Language Audio Anti-Spoofing Datasetcited by 56
  2. 2026[b] = [d] - [t] + [p]: Self-supervised Speech Models Discover Phonological Vector Arithmeticcited by 12
  3. 2026Self-Supervised Speech Models Encode Phonetic Context via Position-dependent Orthogonal Subspacescited by 8
  4. 2025Improving Cross-Lingual Phonetic Representation of Low-Resource Languages Through Language Similarity Analysiscited by 5
  5. 2024Are audio DeepFake detection models polyglots?cited by 3
  6. 2025SEA-Spoof: Bridging The Gap in Multilingual Audio Deepfake Detection for South-East Asiancited by 2
  7. 2026Language Orthogonalization of Self-Supervised Speech Representations for Cross-lingual Parkinson's Detectioncited by 2
  8. 2025How Far Do SSL Speech Models Listen for Tone? Temporal Focus of Tone Representation under Low-resource Transfercited by 1
  9. 2026When Spoof Detectors Travel: Evaluation Across 66 Languages in the Low-Resource Language Spoofing Corpuscited by 1
  10. 2026Scaling Self-Supervised Speech Models Uncovers Deep Linguistic Relationships: Evidence from the Pacific Clustercited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Mitigating Accent-Language Confusion in Self-Supervised Speech Representations for Language Identificationcited by 0