Does Current Deepfake Audio Detection Model Effectively Detect ALM-based Deepfake Audio?

TL;DR

This study evaluates the effectiveness of current deepfake audio detection models against ALM-based deepfake audio, finding that recent codecs trained detection methods perform exceptionally well, achieving near-perfect accuracy.

Contribution

The paper provides the first comprehensive evaluation of detection models specifically against ALM-based deepfake audio using recent datasets and methods.

Findings

Codec-trained detection models achieve 0% error rate on most ALM tests.

Current detection methods are highly effective against ALM-based deepfake audio.

Results suggest promising future research directions in this area.

Abstract

Currently, Audio Language Models (ALMs) are rapidly advancing due to the developments in large language models and audio neural codecs. These ALMs have significantly lowered the barrier to creating deepfake audio, generating highly realistic and diverse types of deepfake audio, which pose severe threats to society. Consequently, effective audio deepfake detection technologies to detect ALM-based audio have become increasingly critical. This paper investigate the effectiveness of current countermeasure (CM) against ALM-based audio. Specifically, we collect 12 types of the latest ALM-based deepfake audio and utilizing the latest CMs to evaluate. Our findings reveal that the latest codec-trained CM can effectively detect ALM-based audio, achieving 0% equal error rate under most ALM test conditions, which exceeded our expectations. This indicates promising directions for future research in ALM-based deepfake audio detection.