MusicDET: Zero-Shot AI-Generated Music Detection

TL;DR

MusicDET introduces a zero-shot, generator-agnostic framework for detecting AI-generated music by modeling real music features with frequency-guided normalizing flows, outperforming traditional methods especially on unseen generators.

Contribution

We propose MusicDET, a novel zero-shot detection method that models real music distributions to identify AI-generated music without prior generated samples.

Findings

MusicDET outperforms traditional discriminative detectors on unseen generators.

The framework effectively models real music features using frequency-guided normalizing flows.

Experiments demonstrate robust detection across multiple datasets.

Abstract

Detecting AI-generated music is crucial for preserving artistic authenticity and preventing the misuse of generative music technologies. However, existing discriminative detectors typically rely on generated samples during training and often suffer from severe performance degradation when confronted with music produced by unseen generators, which limits their real-world applicability. To address this issue, we formulate a zero-shot setting for AI-generated music detection, where the detector is trained exclusively on real music without access to any generated samples. Under this setting, we propose MusicDET, a generator-agnostic detection framework based on frequency-guided normalizing flows that probabilistically models the distribution of real music features. By evaluating the likelihood of an input sample under the learned real-music distribution, MusicDET enables effective detection of out-of-distribution music signals. Experiments on the FakeMusicCaps and SONICS datasets show that MusicDET consistently outperforms conventional discriminative detectors, particularly when detecting music generated by previously unseen models.