Incorporating intentional and emotional behaviors into a virtual human for better customer-engineer-interaction

Providing customer support for technical products means an essential effort for enterprises to satisfy the customer's needs and to challenge rivals in business. This paper introduces a Virtual Human Framework for a better Customer-Engineer-Interaction. We put emphasis on a preferably natural conversation achieved by continuously analyzing behaviors and emotions of the human user, suggesting his or her intentions and diversification of active and passive intentional behaviors. The underlying architecture is an extension to the Generic Embodied Conversational Agent Framework which was developed to ease the integration of heterogeneous components into an Embodied Conversational Agent system. These extensions are mainly influenced by SAIBA's architecture for a Multimodal Behavior Generation Framework. Although the system has only been accomplished to about 50% partial results show that our approach has the potential to create a more natural-like conversational situation.