Generating Socio‐emotional Behaviors
Chloé Clavel · 2019
This chapter explores verbal content generation for agents to consider the way in which the prosody of an utterance is generated as a function of the dialog acts and social attitudes that are being expressed. This relates to the research question relating to the means of modeling sequences of prosodic events in a social attitude generation system for an embodied conversation agent. The chapter extends the research question to a multimodal context, proposing an innovative model for generating co-verbal gestures based on verbal and prosodic content. The computational models presented here build on literature in the fields of conversational analysis and psycholinguistics, and on corpus analysis. In the case of the co-verbal gesture generation model, only qualitative analysis was used. To generate the socio-emotional behaviors, the authors tested sequence mining methods as a means of reducing the effort needed to move from corpus analysis to a computational model.