Multimodal Joke Generation and Paralinguistic Personalization for a Socially-Aware Robot

Abstract

Robot humor is typically scripted by the human. This work presents a socially-aware robot which generates multimodal jokes for use in real-time human-robot dialogs, including appropriate prosody and non-verbal behaviors. It personalizes the paralinguistic presentation strategy based on socially-aware reinforcement learning, which interprets human social signals and aims to maximize user amusement.