Latent-Mark: An Audio Watermark Robust to Neural Resynthesis

• URL: http://arxiv.org/abs/2603.05310v1
• Date: Thu, 05 Mar 2026 15:51:09 GMT
• Title: Latent-Mark: An Audio Watermark Robust to Neural Resynthesis
• Authors: Yen-Shan Chen, Shih-Yu Lai, Ying-Jung Tsou, Yi-Cheng Lin, Bing-Yu Chen, Yun-Nung Chen, Hung-Yi Lee, Shang-Tse Chen,
• Abstract summary: Latent-Mark is the first zero-bit audio watermarking framework designed to survive semantic compression.<n>Our key insight is that robustness to the encode-decode process requires embedding the watermark within the invariant latent space.<n>Our work inspires future research into universal watermarking frameworks capable of maintaining integrity across increasingly complex and diverse generative distortions.
• Score: 62.09761127079914
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• 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.

Related papers

• SKeDA: A Generative Watermarking Framework for Text-to-video Diffusion Models [40.540302276054376]
  We propose a generative watermarking framework tailored for text-to-video diffusion models.<n> SKeDA consists of two components: (1) Shuffle-Key-based Distribution-preserving Sampling (SKe) employs a single base pseudo-random binary sequence for watermark encryption and derives frame-level encryption sequences through permutation.<n>Extensive experiments demonstrate that SKeDA preserves high video generation quality and watermark robustness.
  arXiv  Detail & Related papers  (2026-02-27T06:18:03Z)
• VocBulwark: Towards Practical Generative Speech Watermarking via Additional-Parameter Injection [10.244226665349483]
  VocBulwark is a framework that freezes generative model parameters to preserve perceptual quality.<n>VocBulwark achieves high-capacity and high-fidelity watermarking, offering robust defense against complex practical scenarios.
  arXiv  Detail & Related papers  (2026-01-30T04:51:50Z)
• T2SMark: Balancing Robustness and Diversity in Noise-as-Watermark for Diffusion Models [89.29541056113442]
  T2SMark is a two-stage watermarking scheme based on Tail-Truncated Sampling (TTS)<n>We evaluate T2SMark on diffusion models with both U-Net and DiT backbones.
  arXiv  Detail & Related papers  (2025-10-25T16:55:55Z)
• MARS-Sep: Multimodal-Aligned Reinforced Sound Separation [72.85468563236005]
  MARS-Sep is a reinforcement learning framework for sound separation.<n>It learns a factorized Beta mask policy that is optimized by a clipped trust-region surrogate.<n>Experiments on multiple benchmarks demonstrate consistent gains in Text-, Audio-, and Image-Queried separation.
  arXiv  Detail & Related papers  (2025-10-12T09:05:28Z)
• StableGuard: Towards Unified Copyright Protection and Tamper Localization in Latent Diffusion Models [55.05404953041403]
  We propose a novel framework that seamlessly integrates a binary watermark into the diffusion generation process.<n>We show that StableGuard consistently outperforms state-of-the-art methods in image fidelity, watermark verification, and tampering localization.
  arXiv  Detail & Related papers  (2025-09-22T16:35:19Z)
• SyncGuard: Robust Audio Watermarking Capable of Countering Desynchronization Attacks [41.25345809241139]
  We propose a learning-based scheme named SyncGuard to address these challenges.<n>Specifically, we design a frame-wise broadcast embedding strategy to embed the watermark in arbitrary-length audio.<n>To further enhance robustness, we introduce a meticulously designed distortion layer.
  arXiv  Detail & Related papers  (2025-08-23T19:28:04Z)
• SecoustiCodec: Cross-Modal Aligned Streaming Single-Codecbook Speech Codec [83.61175662066364]
  Speech codecs serve as a crucial bridge in unifying speech and text language models.<n>Existing methods face several challenges in semantic encoding.<n>We propose SecoustiCodec, a cross-modal aligned low-bitrate streaming speech codecs.
  arXiv  Detail & Related papers  (2025-08-04T19:22:14Z)
• Towards Generalized Source Tracing for Codec-Based Deepfake Speech [52.68106957822706]
  We introduce the Semantic-Acoustic Source Tracing Network (SASTNet), which jointly leverages Whisper for semantic feature encoding and Wav2vec2 with AudioMAE for acoustic feature encoding.<n>Our proposed SASTNet achieves state-of-the-art performance on the CoSG test set of the CodecFake+ dataset, demonstrating its effectiveness for reliable source tracing.
  arXiv  Detail & Related papers  (2025-06-08T21:36:10Z)
• Video Signature: In-generation Watermarking for Latent Video Diffusion Models [42.064769031646904]
  Video Signature (VID SIG) is an in-generation watermarking method for latent video diffusion models.<n>We achieve this by partially fine-tuning the latent decoder, where Perturbation-Aware Suppression (PAS) pre-identifies and freezes perceptually sensitive layers.<n> Experimental results show that VID SIG achieves the best overall performance in watermark extraction, visual quality, and generation efficiency.
  arXiv  Detail & Related papers  (2025-05-31T17:43:54Z)
• Enhancing Quantum Dense Coding Robustness Using Information Entropy-Based Metrics [0.0]
  Superdense Coding is a cornerstone in secure quantum communication, exploiting pre-shared entanglement to encode two classical bits within a single qubit.<n>Traditional methods, such as error correcting codes or entanglement distillation, are generally inadequate for dynamically varying noise conditions.<n>This work introduces an adaptive protocol that integrates the five-qubit perfect code with a novel global adaptive purification that avoids discarding entangled pairs.
  arXiv  Detail & Related papers  (2025-04-17T01:29:40Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.