Multimodal animation control

Abstract

In this thesis, we present a multimodal animation control system. Our approach is based on a human-centric computing model proposed by Project Oxygen at MIT Laboratory for Computer Science. Our system allows the user to create and control animation in real time using the speech interface developed using SpeechBuilder. The user can also fall back to traditional input modes should the speech interface fail. We assume that the user has no prior knowledge and experience in animation and yet enable him to create interesting and meaningful animation naturally and fluently. We argue that our system can be used in a number of applications ranging from PowerPoint presentations to simulations to children's storytelling tools.