AWARE: Audio Watermarking with Adversarial Resistance to Edits

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Pavlović, Kosta, Stanarević, Lazar, Nedić, Petar, Kovačević, Elena Nešović Slavko, Djurović, Igor
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866908888853381120
author Pavlović, Kosta
Stanarević, Lazar
Nedić, Petar
Kovačević, Elena Nešović Slavko
Djurović, Igor
author_facet Pavlović, Kosta
Stanarević, Lazar
Nedić, Petar
Kovačević, Elena Nešović Slavko
Djurović, Igor
contents Prevailing practice in learning-based audio watermarking is to pursue robustness by expanding the set of simulated distortions during training. However, such surrogates are narrow and prone to overfitting. This paper presents AWARE (Audio Watermarking with Adversarial Resistance to Edits), an alternative approach that avoids reliance on attack-simulation stacks and handcrafted differentiable distortions. Embedding is obtained through adversarial optimization in the time-frequency domain under a level-proportional perceptual budget. Detection employs a time-order-agnostic detector with a Bitwise Readout Head (BRH) that aggregates temporal evidence into one score per watermark bit, enabling reliable watermark decoding even under desynchronization and temporal cuts. Empirically, AWARE attains high audio quality and speech intelligibility (PESQ/STOI) and consistently low BER across various audio edits, often surpassing representative state-of-the-art learning-based systems.
format Preprint
id arxiv_https___arxiv_org_abs_2510_17512
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle AWARE: Audio Watermarking with Adversarial Resistance to Edits
Pavlović, Kosta
Stanarević, Lazar
Nedić, Petar
Kovačević, Elena Nešović Slavko
Djurović, Igor
Sound
Machine Learning
Multimedia
Audio and Speech Processing
Prevailing practice in learning-based audio watermarking is to pursue robustness by expanding the set of simulated distortions during training. However, such surrogates are narrow and prone to overfitting. This paper presents AWARE (Audio Watermarking with Adversarial Resistance to Edits), an alternative approach that avoids reliance on attack-simulation stacks and handcrafted differentiable distortions. Embedding is obtained through adversarial optimization in the time-frequency domain under a level-proportional perceptual budget. Detection employs a time-order-agnostic detector with a Bitwise Readout Head (BRH) that aggregates temporal evidence into one score per watermark bit, enabling reliable watermark decoding even under desynchronization and temporal cuts. Empirically, AWARE attains high audio quality and speech intelligibility (PESQ/STOI) and consistently low BER across various audio edits, often surpassing representative state-of-the-art learning-based systems.
title AWARE: Audio Watermarking with Adversarial Resistance to Edits
topic Sound
Machine Learning
Multimedia
Audio and Speech Processing
url https://arxiv.org/abs/2510.17512