Note: The code for this project is currently being prepared for release.
This repository will contain the official implementation for the paper:
"Uncertainty-Guided Domain Augmentation for Domain Generalization in Speaker Verification and Anti-Spoofing"
Authors: Jin Li, Man-Wai Mak, Kong Aik Lee, Oldลich Plchot
Accepted by Proc. ICASSP, 2026
We are currently finalizing the code cleanup and documentation. The full source code will be released as soon as we finish it.
[Automatic speaker verification and audio anti-spoofing perform unreliably under domain mismatch conditions, such as a mismatch in channels and recording devices. Domain generalization aims to ensure reliable performance on unseen conditions while maintaining high discrimination on seen domains. This paper proposes a novel method UDANet (Uncertainty-guided Domain Augmentation Network) for domain generalization. The key idea of UDANet is to augment the source domain in neural representation through the Uncertainty-guided Mock Domain Generator (UMDG). The UMDG uses a Gaussian mixture density network to learn class features that are robust across domains. This uncertainty guides domain augmentation toward ambiguous representations, generating more diverse and challenging domain samples. By augmenting uncertain neural representations, this approach expands the modelโs distributional space and enhances robustness across different domains. Theoretical analyses prove that UMDG matches target shift moments to upper-bound target risk, guaranteeing a more domain robust model. Experimental results demonstrate that UDANet improves performance on standard benchmarks, especially in cross-database and challenging spoofing scenarios.]
Please Star โญ this repository to be notified when the code is released.