arXiv：cs.AI· Yihang Zhang, Weicai Yan, Qifei Zhang, Yangyang Wu, Tao Jin·· 6 小时前
GAMW：面向 TTS 溯源、注入生成过程的 Mel 域水印框架
GAMW: Generation-Aware Mel-Domain Watermarking for TTS Traceability
AI 导读
GAMW 是一种集成到 TTS 生成过程的水印注入框架，在波形合成前将内容自适应的多比特水印残差嵌入生成表征，解码目标结合 focal modulation 与动态逐比特加权。在 16-bit 载荷下，GAMW 在单一扰动和复合扰动下分别比最强基线提升 9.2 和 23.5 个百分点的解码准确率，同时保持有竞争力的语音质量，更高载荷下仍可可靠解码干净语音。代码已开源。
正文
Abstract:As text-to-speech (TTS) systems continue to advance, provenance tracking and security verification of synthesized speech have become increasingly important. Audio watermarking provides active provenance by embedding imperceptible, machine-recoverable information into speech. However, existing post-hoc methods typically modify the waveform after synthesis, leaving the watermark vulnerable to compound perturbations. To address this limitation, we propose GAMW, a watermark injection framework integrated into the TTS generation process. Before waveform synthesis, GAMW embeds a content-adaptive multi-bit watermark residual into the generated representation, and its decoding objective combines focal modulation with dynamic per-bit weighting. We train and evaluate GAMW separately on two TTS backbones. With a 16-bit payload, GAMW improves decoding accuracy over the strongest baseline by 9.2 and 23.5 percentage points under individual and compound perturbations, respectively, while maintaining competitive speech quality. Reliable clean-speech decoding is also retained at higher payloads. Code is available at this https URL.
| Subjects: | Sound (cs.SD); Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR) |
| Cite as: | arXiv:2610.15082 [cs.SD] |
| (or arXiv:2610.15082v1 [cs.SD] for this version) | |
| https://doi.org/10.48550/arXiv.2610.15082 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Yihang Zhang [view email]
[v1]
Sat, 10 Oct 2026 17:30:02 UTC (459 KB)
来源：arXiv：cs.AI · arxiv.org