Multi-modal User Interface Design for a Face and Voice Recognition Biometric Authentication System

Biometrics refer to unique measurable characteristics and information regarding individual's health, physical or mental condition and can be used to uniquely authenticate or verify a person's identity. They can be sorted in physiological such as fingerprints, palm print, face recognition, iris recognition, retina and DNA and behavioral such as typing rhythm (i.e. signature) and voice and can be described based on the uniqueness, potential change with time (i.e. facial changes), the feasibility to be collected (i.e. fingerprints) and the purposes of usage. In this work we study the use of a biometric technology for eHealth. We present the SpeechXRays project initiative that aims to provide a solution combining the convenience and cost-effectiveness of face and voice biometrics, achieving better accuracies by combining it with video, and bringing superior anti-spoofing capabilities. We explain how a novel user interface biometric platform is designed and adapted, for an eHealth use case, to enable secure access for medical specialists, nurses and patients to a collaborative eHealth platform that provides access to clinical and health related data within and possible outside a hospital. This is the first study, in the field, that gathered all necessary requirements (for a voice/face biometric system) and provides a formative evaluation and implementation of the SpeechXrays system user interface, for both end users and administrators, following a user-centered design approach, based on the holistic consideration of the user experience and the technical implication and functional requirements of the platform.