TTS speech watermarks improve traceability under signal changes
GAMW: Generation-Aware Mel-Domain Watermarking for TTS Traceability
SoundArtificial IntelligenceCryptography and Security
Summary
Text-to-speech systems create spoken audio from text, but it can be hard to track who made the speech and verify its origin. The authors propose a method called GAMW that hides a secret digital code inside the speech as it is made, rather than adding it afterward. This makes the hidden code more reliable even if the audio is altered or affected by noise. Their approach improves the accuracy of detecting this code while keeping the speech quality natural.
What this means in practice
- •For voice security teams: Embed robust watermarks inside TTS audio to verify its source even after signal distortions or edits.
- •For voice assistant developers: Integrate watermarking during synthesis for traceable synthetic speech that resists common audio changes.
Authors
Yihang Zhang, Weicai Yan, Qifei Zhang, Yangyang Wu, Tao Jin
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 https://github.com/ZConJugaTe/GAMW.