Conception et Evaluation d'un Modèle d'Expressivité pour les Gestes des Agents Conversationnels

Stéphanie Buisine (1) , Björn Hartmann (2) , Maurizio Mancini (3) , Catherine Pelachaud (4)

1 LCPI - Laboratoire Conception de Produits et Innovation
2 Computer Science Department [Stanford]
3 IUT Montreuil - Institut universitaire technologique de Montreuil
4 LTCI - Laboratoire Traitement et Communication de l'Information

To increase the believability and life-likeness of Embodied Conversational Agents, we introduce a behavior synthesis model for the generation of expressive gesturing. A small set of dimensions of expressivity is used to characterize individual variability of movement. We empirically evaluate our implementation in two separate user studies. The results suggest that our approach works well for a subset of expressive behavior. However, animation fidelity is not high enough to realize subtle changes. Interaction effects between different parameters need to be studied further.

Domaines: Interface homme-machine [cs.HC]

Stéphanie Buisine, Björn Hartmann, Maurizio Mancini, Catherine Pelachaud. Conception et Evaluation d'un Modèle d'Expressivité pour les Gestes des Agents Conversationnels. Revue des Sciences et Technologies de l'Information - Série RIA : Revue d'Intelligence Artificielle, 2006, pp.621-638. ⟨hal-00786518⟩