Saved in:
Bibliographic Details
Main Authors: Yuan, Zheqi, Huang, Yucheng, Sun, Guangzhi, Jin, Zengrui, Zhang, Chao
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2511.02278
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910047801442304
author Yuan, Zheqi
Huang, Yucheng
Sun, Guangzhi
Jin, Zengrui
Zhang, Chao
author_facet Yuan, Zheqi
Huang, Yucheng
Sun, Guangzhi
Jin, Zengrui
Zhang, Chao
contents Audio watermarking is essential for verifying speech authenticity, yet single-watermark schemes often struggle against sophisticated distortions such as neural reconstruction and adversarial attacks. To address this limitation, we introduce a multiplexing paradigm that combines multiple watermarking techniques to leverage their inherent complementarities. We explore both parallel and sequential multiplexing strategies and propose perceptual-adaptive time-frequency multiplexing (PA-TFM), a robust training-free approach. To further enhance performance, we introduce MaskNet, a novel model-based framework designed to learn effective time-domain multiplexing. Experimental results on the LibriSpeech and Common Voice datasets under 14 diverse attack types, including high-strength white-box and neural reconstruction attacks, demonstrate that both PA-TFM and MaskNet considerably outperform existing single-watermark baselines, establishing a resilient paradigm for real-world audio protection.
format Preprint
id arxiv_https___arxiv_org_abs_2511_02278
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Multiplexing Neural Audio Watermarks
Yuan, Zheqi
Huang, Yucheng
Sun, Guangzhi
Jin, Zengrui
Zhang, Chao
Audio and Speech Processing
Audio watermarking is essential for verifying speech authenticity, yet single-watermark schemes often struggle against sophisticated distortions such as neural reconstruction and adversarial attacks. To address this limitation, we introduce a multiplexing paradigm that combines multiple watermarking techniques to leverage their inherent complementarities. We explore both parallel and sequential multiplexing strategies and propose perceptual-adaptive time-frequency multiplexing (PA-TFM), a robust training-free approach. To further enhance performance, we introduce MaskNet, a novel model-based framework designed to learn effective time-domain multiplexing. Experimental results on the LibriSpeech and Common Voice datasets under 14 diverse attack types, including high-strength white-box and neural reconstruction attacks, demonstrate that both PA-TFM and MaskNet considerably outperform existing single-watermark baselines, establishing a resilient paradigm for real-world audio protection.
title Multiplexing Neural Audio Watermarks
topic Audio and Speech Processing
url https://arxiv.org/abs/2511.02278