Designing Speech with Computational Linguistics for a Virtual Medical Assistant Using Situational Leadership
Résumé
In emergency medical procedures, positive and trusting interaction between followers and leaders are imperative. That interaction is even more important when a virtual agent assumes the leader role and a human assumes the follower role. In order to manage the human-computer interaction, situational leadership is employed to match the human to an appropriate leadership style embodied by the agent.
This paper explores how different leadership styles can be conveyed by a virtual agent through an analysis of utterances made by doctors and coordinators during emergency simulations. We create a corpus which comprises utterances from simulation videos of medical emergencies. Each utterance is annotated with a leadership style. After analysis involving k-means clustering, we compile easily-reproducible rules that dictate how speech should appear in each leadership style for use in a virtual agent system.