VocBulwark: Towards Practical Generative Speech Watermarking via Additional-Parameter Injection

2026-01-30

AI Summary

This work addresses the dual challenge of preserving high audio fidelity while ensuring robust watermarking against misuse of highly natural synthetic speech. To this end, the authors propose VocBulwark, a framework that embeds watermarks deeply into acoustic features by injecting learnable temporal adapters without modifying the frozen weights of the primary speech generation model. The method jointly optimizes watermark embedding and extraction through a coarse-to-fine gated extractor and an accuracy-guided curriculum learning strategy. VocBulwark achieves high-capacity watermarking with minimal impact on perceptual quality and demonstrates strong robustness against diverse signal processing attacks, including codec compression, resampling, and variable-length manipulations.

Technology Category

• Machine Learning: Adversarial Learning & Robustness
• Natural Language Processing: Speech
• Computer Vision: Adversarial Attacks & Robustness

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Generated speech achieves human-level naturalness but escalates security risks of misuse. However, existing watermarking methods fail to reconcile fidelity with robustness, as they rely either on simple superposition in the noise space or on intrusive alterations to model weights. To bridge this gap, we propose VocBulwark, an additional-parameter injection framework that freezes generative model parameters to preserve perceptual quality. Specifically, we design a Temporal Adapter to deeply entangle watermarks with acoustic attributes, synergizing with a Coarse-to-Fine Gated Extractor to resist advanced attacks. Furthermore, we develop an Accuracy-Guided Optimization Curriculum that dynamically orchestrates gradient flow to resolve the optimization conflict between fidelity and robustness. Comprehensive experiments demonstrate that VocBulwark achieves high-capacity and high-fidelity watermarking, offering robust defense against complex practical scenarios, with resilience to Codec regenerations and variable-length manipulations.

Problem

• speech watermarking
• fidelity
• robustness
• generative speech
• security

Innovation

• speech watermarking
• parameter injection
• temporal adapter
• robustness-fidelity tradeoff
• generative audio security