Deepfakes as a threat to a speaker and facial recognition: an overview of tools and attack vectors

Face deepfakes, Speech deepfakes, Biometrics systems, Facial recognition, Speaker recognition, Deepfake detection, Cybersecurity

Deepfakes present an emerging threat in cyberspace. Recent developments in machine learning make deepfakes highly believable, and very difficult to differentiate between what is real and what is fake. Not only humans but also machines struggle to identify deepfakes. Current speaker and facial recognition systems might be easily fooled by carefully prepared synthetic media - deepfakes. We provide a detailed overview of the state-of-the-art deepfake creation and detection methods for selected visual and audio domains. In contrast to other deepfake surveys, we focus on the threats that deepfakes represent to biometrics systems (e.g., spoofing). We discuss both facial and speechdeepfakes, and for each domain, we define deepfake categories and their differences. For each deepfake category, we provide an overview of available tools for creation, datasets, and detection methods. Our main contribution is a definition of attack vectors concerning the differences between categories and reported real-world attacks to evaluate each category's threats to selected categories of biometrics systems.