Task-Aware Joint Pruning and Distillation for Efficient Audio Deepfake Detection

arXiv:2610.05264 · cs.SD, cs.AI · Submitted 2026-10-04 · Read on arXiv

cs.SD, cs.AI

Submitted: 2026-10-04

Updated: 2026-10-04

Comments: 6 pages, 4 figures, accepted to Interspeech 2026

DOI: 10.21437/Interspeech.2026-1766

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Distillation — 1× in this paper · explained in 2 episodes
• Pruning — 1× in this paper · explained in 4 episodes

Sources

• XLS-R: Self-supervised Cross-lingual Speech Representation Learning at Scale
• Automatic speaker verification spoofing and deepfake detection using wav2vec 2.0 and data augmentation
• Hybrid Pruning: In-Situ Compression of Self-Supervised Speech Models for Speaker Verification and Anti-Spoofing
• ASVspoof 2019: Future Horizons in Spoofed and Fake Audio Detection
• ASVspoof 2021: accelerating progress in spoofed and deepfake speech detection
• ASVspoof 5: Crowdsourced Speech Data, Deepfakes, and Adversarial Attacks at Scale
• What Do Compressed Deep Neural Networks Forget?
• Post-training for Deepfake Speech Detection

Related papers

• Few-Shot Open-Set Audio Classification via Transductive Prototype Refinement and Class Logit Enhancement
• Spectral Masking and Interpolation Attack (SMIA): A Black-box Adversarial Attack against Voice Authentication and Anti-Spoofing Systems
• Remix the Timbre: Diffusion-Based Style Transfer Across Polyphonic Stems
• AVMeme Exam: A Multimodal Multilingual Multicultural Benchmark for LLMs' Contextual and Cultural Knowledge and Thinking
• SoundWeaver: Compositional Warm-Starting for Text-to-Audio Diffusion Serving
• WASIL: In-the-Wild Arabic Spoken Interactions with LLMs