Audio Pirates: Black-box Audio Watermark Removal via Diffusion Priors

2026-05-28

AI Summary

This work addresses the vulnerability of existing audio watermarking schemes under black-box attacks, where current removal methods often degrade audio quality or rely on prior knowledge of the watermarking mechanism. To overcome these limitations, we propose DiffErase, the first approach to leverage diffusion model priors for black-box watermark removal. By perturbing watermarked audio into an intermediate noise layer of a pre-trained diffusion model and reconstructing it through the denoising process, DiffErase effectively suppresses watermark signals without requiring any information about the underlying watermarking algorithm. Experimental results demonstrate that DiffErase significantly reduces watermark detection rates across multiple audio domains while preserving high subjective listening quality and strong objective audio fidelity metrics, thereby revealing a critical vulnerability of contemporary audio watermarking systems to diffusion-based attacks.

Abstract

With the rise of AI-generated audio, watermarking has become widely used for detecting misuse and protecting intellectual property. However, adversaries may try to remove these watermarks, making it critical to evaluate how well watermarking schemes withstand removal attacks. Existing attacks are often impractical: they either noticeably degrade perceptual quality or require access to the watermarking scheme. We propose DiffErase, a black-box watermark removal attack that assumes no knowledge of the target watermarking scheme while maintaining perceptual quality. DiffErase perturbs watermarked audio to an intermediate diffusion noise level and regenerates it using a pretrained denoising model, effectively suppressing watermark signals. Theoretical analysis and extensive experiments demonstrate that inaudible audio watermarks are highly vulnerable: across multiple audio domains, DiffErase consistently removes watermarks while preserving perceptual quality. These findings highlight the need for future audio watermarking designs to consider diffusion-based threats. Code and demos are available at https://differase.github.io/DiffErase/.

Problem

• audio watermarking
• watermark removal
• black-box attack
• perceptual quality
• diffusion models

Innovation

• diffusion priors
• black-box attack
• audio watermark removal
• denoising diffusion model
• perceptual quality preservation