What to Remember: Self-Adaptive Continual Learning for Audio Deepfake Detection

TL;DR

This paper introduces Radian Weight Modification, a continual learning method that improves audio deepfake detection by effectively distinguishing genuine and fake audio classes, reducing forgetting and enhancing knowledge retention.

Contribution

The paper proposes RWM, a novel continual learning approach that categorizes classes based on feature distribution and applies gradient modifications, advancing audio deepfake detection and broader machine learning tasks.

Findings

RWM outperforms existing continual learning methods in deepfake detection.

RWM effectively mitigates forgetting and enhances knowledge acquisition.

Applicable to other domains like image recognition.

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 direction for distinct data types. Experimental evaluations against mainstream continual learning methods reveal the superiority of RWM in terms of knowledge acquisition and mitigating forgetting in audio deepfake detection. Furthermore, RWM's applicability extends beyond audio deepfake detection, demonstrating its potential significance in diverse machine learning domains such as image recognition.