DFALLM: Achieving Generalizable Multitask Deepfake Detection by Optimizing Audio LLM Components

TL;DR

This paper introduces a novel ALLM architecture optimized for generalizable audio deepfake detection, achieving state-of-the-art results across multiple datasets and tasks by carefully selecting model components.

Contribution

It proposes a new ALLM structure that enhances generalization to out-of-domain deepfake detection and related tasks, addressing previous bottlenecks in audio LLM performance.

Findings

Achieves up to 95.76% accuracy on multiple datasets

Outperforms existing models in deepfake attribution and localization

Demonstrates the importance of component selection in ALLMs

Abstract

Audio deepfake detection has recently garnered public concern due to its implications for security and reliability. Traditional deep learning methods have been widely applied to this task but often lack generalisability when confronted with newly emerging spoofing techniques and more tasks such as spoof attribution recognition rather than simple binary classification. In principle, Large Language Models (LLMs) are considered to possess the needed generalisation capabilities. However, previous research on Audio LLMs (ALLMs) indicates a generalization bottleneck in audio deepfake detection performance, even when sufficient data is available. Consequently, this study investigates the model architecture and examines the effects of the primary components of ALLMs, namely the audio encoder and the text-based LLM. Our experiments demonstrate that the careful selection and combination of audio encoders and text-based LLMs are crucial for unlocking the deepfake detection potential of ALLMs. We further propose an ALLM structure capable of generalizing deepfake detection abilities to out-of-domain spoofing tests and other deepfake tasks, such as spoof positioning and spoof attribution recognition. Our proposed model architecture achieves state-of-the-art (SOTA) performance across multiple datasets, including ASVSpoof2019, InTheWild, and Demopage, with accuracy reaching up to 95.76% on average, and exhibits competitive capabilities in other deepfake detection tasks such as attribution, and localisation compared to SOTA audio understanding models. Data and codes are provided in supplementary materials.