Traceback Translators Against Forgetting in Continual Fake Speech Detection
arXiv:2607.12569
The paper proposes a method that uses a domain‑translator network to map new fake‑speech data back into the feature space of an existing detector, allowing continual learning without forgetting previously learned detection capabilities.
Abstract
Fake speech detectors are increasingly challenged by the development of new and more accurate generative models. To cope with this problem, continual learning techniques are nowadays widely considered feasible strategies for updating models to new datasets, but they also lead to decreased performance on previously seen samples (catastrophic forgetting). In this work, we propose a forgetting-resilient solution based on the adoption of domain translators within a frozen detector, which remaps the new feature spaces into the original ones by means of a traceback translator network. Experimental results show that this strategy enables the achievement of high detection rates with respect to traditional retraining, while minimizing the computational effort and preserving the detection accuracy on previous data.
Accepted at EUSIPCO 2026