Training-Free Deepfake Voice Recognition by Leveraging Large-Scale Pre-Trained Models

arXiv:2405.02179v1 Announce Type: cross
Abstract: Generalization is a main issue for current audio deepfake detectors, which struggle to provide reliable results on out-of-distribution data. Given the speed at which more and more accurate synthesis methods are developed, it is very important to design techniques that work well also on data they were not trained for.In this paper we study the potential of large-scale pre-trained models for audio deepfake detection, with special focus on generalization ability. To this end, the detection …

abstract arxiv audio audio deepfake cs.cv cs.sd current data deepfake deepfake detectors design detectors distribution eess.as free issue pre-trained models recognition results scale speed struggle synthesis training type voice voice recognition work