Expressive Multimodal Conversational Acts for SAIBA Agents

Jérémy Rivière (1) , Carole Adam (1) , Sylvie Pesty (1) , Catherine Pelachaud (2, 3) , Nadine Guiraud (4, 5) , Dominique Longin (4, 3) , Emiliano Lorini (4, 3)

1 LIG - Laboratoire d'Informatique de Grenoble
2 LTCI - Laboratoire Traitement et Communication de l'Information
3 CNRS - Centre National de la Recherche Scientifique
4 IRIT-LILaC - Logique, Interaction, Langue et Calcul
5 UT3 - Université Toulouse III - Paul Sabatier

We discuss here the need to define what we call an agent con-versational language, a language for Embodied Conversational Agents (ECA) to have conversations with a human. We propose a set of Ex-pressive Multimodal Conversation Acts (EMCA), which is based on the Expressive Speech Acts that we introduced in a previous work, enriched with the multimodal expression of the emotions linked to these acts. We have then implemented these EMCA for SAIBA-compliant ECA, and specifically for Greta. We were able to use Greta in experiments aimed at assessing the benefits of our language in terms of perceived sincerity and believability of an ECA using it to interact with a human user.

Keywords: ECA; Emotions; Interaction; Dialogue

Domains: Artificial Intelligence [cs.AI]; Logic in Computer Science [cs.LO]; Multiagent Systems [cs.MA]

Jérémy Rivière, Carole Adam, Sylvie Pesty, Catherine Pelachaud, Nadine Guiraud, et al.. Expressive Multimodal Conversational Acts for SAIBA Agents. 11th International Workshop on Intelligent Virtual Agents (IVA 2011), Sep 2011, Reykjavik, Iceland. pp.316-323, ⟨10.1007/978-3-642-23974-8_34⟩. ⟨hal-00950872⟩