Latent-Mark: An Audio Watermark Robust to Neural Resynthesis

TL;DR

Latent-Mark introduces a novel audio watermarking method that embeds watermarks into the invariant latent space of neural codecs, achieving robustness against neural resynthesis and traditional DSP attacks while maintaining imperceptibility.

Contribution

It is the first zero-bit audio watermarking framework designed to survive semantic compression by embedding in the codec's invariant latent space.

Findings

Robust to neural resynthesis and DSP attacks

Achieves state-of-the-art resilience while maintaining perceptual quality

Effective transferability to unseen neural codecs

Abstract

While existing audio watermarking techniques have achieved strong robustness against traditional digital signal processing (DSP) attacks, they remain vulnerable to neural resynthesis. This occurs because modern neural audio codecs act as semantic filters and discard the imperceptible waveform variations used in prior watermarking methods. To address this limitation, we propose Latent-Mark, the first zero-bit audio watermarking framework designed to survive semantic compression. Our key insight is that robustness to the encode-decode process requires embedding the watermark within the codec's invariant latent space. We achieve this by optimizing the audio waveform to induce a detectable directional shift in its encoded latent representation, while constraining perturbations to align with the natural audio manifold to ensure imperceptibility. To prevent overfitting to a single codec's quantization rules, we introduce Cross-Codec Optimization, jointly optimizing the waveform across multiple surrogate codecs to target shared latent invariants. Extensive evaluations demonstrate robust zero-shot transferability to unseen neural codecs, achieving state-of-the-art resilience against traditional DSP attacks while preserving perceptual imperceptibility. Our work inspires future research into universal watermarking frameworks capable of maintaining integrity across increasingly complex and diverse generative distortions.