Audio Codec Augmentation for Robust Collaborative Watermarking of Speech Synthesis

TL;DR

This paper enhances collaborative audio watermarking techniques to improve robustness against various audio codecs, ensuring reliable detection of synthetic speech with minimal perceptual quality loss.

Contribution

It extends channel augmentation methods to non-differentiable codecs using waveform estimators and demonstrates transferability across different codec types and bitrates.

Findings

Robust watermarking can be achieved with black-box audio codecs.

Channel augmentation transfers well to traditional codecs.

Negligible perceptual degradation at high bitrates.

Abstract

Automatic detection of synthetic speech is becoming increasingly important as current synthesis methods are both near indistinguishable from human speech and widely accessible to the public. Audio watermarking and other active disclosure methods of are attracting research activity, as they can complement traditional deepfake defenses based on passive detection. In both active and passive detection, robustness is of major interest. Traditional audio watermarks are particularly susceptible to removal attacks by audio codec application. Most generated speech and audio content released into the wild passes through an audio codec purely as a distribution method. We recently proposed collaborative watermarking as method for making generated speech more easily detectable over a noisy but differentiable transmission channel. This paper extends the channel augmentation to work with non-differentiable traditional audio codecs and neural audio codecs and evaluates transferability and effect of codec bitrate over various configurations. The results show that collaborative watermarking can be reliably augmented by black-box audio codecs using a waveform-domain straight-through-estimator for gradient approximation. Furthermore, that results show that channel augmentation with a neural audio codec transfers well to traditional codecs. Listening tests demonstrate collaborative watermarking incurs negligible perceptual degradation with high bitrate codecs or DAC at 8kbps.