Description
Date depot: 2 mars 2020
Titre: Front-End Feature Optimization for Anti-spoofing
Directeur de thèse:
Nicholas EVANS (Eurecom)
Domaine scientifique: Sciences et technologies de l'information et de la communication
Thématique CNRS : Non defini
Resumé:
Automatic speaker verification (ASV) technology has matured over recent years to become a low-cost and reliable approach to person recognition. Unfortunately, however, and as is true for all biometric modalities, concerns regarding security vulnerabilities can still form a barrier to exploitation. Vulnerabilities to spoofing, also known as presentation attacks, are one example spoofing refers to the manipulation of a biometric system by a fraudster impersonating another enrolled person. For medium to high security applications, such vulnerabilities are clearly unacceptable.
Doctorant.e: Tak Hemlata