A Novel Unity-based Realizer for the Realization of Conversational Behavior on Embodied Conversational Agents

Abstract: Embodied conversational agents are virtual entities that tend to imitate as many features of face-face dialogs as possible. In order to achieve this goal, the ability to reproduce synchronized verbal and co-verbal signals coupled into conversational behavior becomes essential. Further, signals such as social cues, attitude (emotions), personality, eye-blinks, and spontaneous head movement are equally important. Modern 3D environments and 3D modeling tools, such as: Maya, Daz3D, Blender, Panda3D and Unity have opened up a completely new possibilities to design virtual entities, which appear almost (if no completely) like real-life persons. However, the modern 3D technology is not designed to handle highly dynamic and interchangeable contexts such as human interaction. Therefore, mostly animations are prepared in advance and support limited diversity as well as limited capacity to adapt to a new set of parameters. In this paper EVA realizer engine, which is a part of proprietary behavior realization components of EVA Framework, is presented. The represented engine is based on Unity game engine. EVA realizer exploits benefits of modern game engines as well as extend them with requirements of co-verbal realizers, by providing interpreter and manager for dynamic and in real-time generated animation. The animation is created and modeled by proprietary external co-verbal behavior generator component.