Spoofing Aware Speaker Verification

Spoofing-aware speaker verification (SASV) aims to build robust speaker recognition systems that are resistant to spoofing attacks, such as deepfakes and voice replays. Current research focuses on integrating automatic speaker verification (ASV) and spoofing countermeasure (CM) systems, often employing techniques like score or embedding fusion with deep neural networks (DNNs), including architectures such as ResNet, ECAPA-TDNN, and WavLM, and exploring novel loss functions like a-DCF. This field is crucial for enhancing the security of voice-based authentication systems and has significant implications for various applications, from security and access control to forensic science.