What To Remember: Self-adaptive Continual Learning For Audio Deepfake Detection

Abstract

The rapid evolution of speech synthesis and voice conversion has raised substantial concerns due to the potential misuse of such technology, prompting a pressing need for effective audio deepfake detection mechanisms. Existing detection models have shown remarkable success in discriminating known deepfake audio, but struggle when encountering new attack types. To address this challenge, one of the emergent effective approaches is continual learning. In this paper, we propose a continual learning approach called Radian Weight Modification (RWM) for audio deepfake detection. The fundamental concept underlying RWM involves categorizing all classes into two groups: those with compact feature distributions across tasks, such as genuine audio, and those with more spread-out distributions, like various types of fake audio. These distinctions are quantified by means of the in-class cosine distance, which subsequently serves as the basis for RWM to introduce a trainable gradient modification di

Stats

• citations24
• S2 citations—
• github stars0
• HF likes0
• heat score10.48
• arxiv keyzhang2023what

Related papers

• Towards Robust Audio Deepfake Detection: A Evolving Benchmark For Continual Learning (2024)0.00
• Region-based Optimization In Continual Learning For Audio Deepfake Detection (2024)4.49
• Continual Learning For Fake Audio Detection (2021)11.49
• Adaptive Re-calibration Of Channel-wise Features For Adversarial Audio Classification (2022)0.00
• The Codecfake Dataset And Countermeasures For The Universally Detection Of Deepfake Audio (2024)10.97
• Defense Against Synthetic Speech: Real-time Detection Of RVC Voice Conversion Attacks (2025)0.00
• Securing Voice Biometrics: One-shot Learning Approach For Audio Deepfake Detection (2023)9.03
• Pitch Imperfect: Detecting Audio Deepfakes Through Acoustic Prosodic Analysis (2025)0.00