MagLive: Near-Field Magnetic Sensing-Based Voice Liveness Detection on Smartphones

Voice authentication has been widely used on smartphones. However, it remains vulnerable to spoofing attacks, where the attacker replays recorded voice samples from authentic humans using loudspeakers to bypass the voice authentication system. In this paper, we present MagLive, a robust voice liveness detection scheme designed for smartphones to mitigate such spoofing attacks. MagLive leverages differences in magnetic field patterns generated by different speakers (i.e., humans or loudspeakers) when speaking for liveness detection. It uses the built-in magnetometer on smartphones to capture these magnetic field changes. Specifically, MagLive utilizes two CNN-based submodels and a self-attention-based feature fusion model to extract effective and robust features. Supervised contrastive learning is then employed to achieve user-irrelevance, device-irrelevance, and content-irrelevance. MagLive imposes no additional burdens on users and does not rely on active sensing or extra devices. We conducted comprehensive experiments with various settings to evaluate the security and robustness of MagLive. Our results demonstrate that MagLive effectively distinguishes between humans and attackers (i.e., loudspeakers), achieving a balanced accuracy of 99.01 of 0.77