GROOT: Generating Robust Watermark for Diffusion-Model-Based Audio Synthesis

TL;DR

GROOT introduces a novel watermarking method for diffusion-model-based audio synthesis, enabling proactive supervision of generated audio with high robustness and minimal impact on audio quality.

Contribution

This paper pioneers a simultaneous watermark generation and audio synthesis paradigm using fixed diffusion models and a dedicated encoder, enhancing robustness over existing methods.

Findings

GROOT achieves around 95% watermark extraction accuracy under compound attacks.

It outperforms state-of-the-art watermarking techniques in robustness.

The method maintains high audio quality with minimal distortion.

Abstract

Amid the burgeoning development of generative models like diffusion models, the task of differentiating synthesized audio from its natural counterpart grows more daunting. Deepfake detection offers a viable solution to combat this challenge. Yet, this defensive measure unintentionally fuels the continued refinement of generative models. Watermarking emerges as a proactive and sustainable tactic, preemptively regulating the creation and dissemination of synthesized content. Thus, this paper, as a pioneer, proposes the generative robust audio watermarking method (Groot), presenting a paradigm for proactively supervising the synthesized audio and its source diffusion models. In this paradigm, the processes of watermark generation and audio synthesis occur simultaneously, facilitated by parameter-fixed diffusion models equipped with a dedicated encoder. The watermark embedded within the audio can subsequently be retrieved by a lightweight decoder. The experimental results highlight Groot's outstanding performance, particularly in terms of robustness, surpassing that of the leading state-of-the-art methods. Beyond its impressive resilience against individual post-processing attacks, Groot exhibits exceptional robustness when facing compound attacks, maintaining an average watermark extraction accuracy of around 95%.