Emotion-Driven Facial Animation for Chat-Bots

Abstract

Facial animation is quickly becoming an important feature of virtual assistants and is gaining traction as the preferred communication technique between man and machine. In addition to lip movement, facial expressions such as those of the eyes and cheeks help in conveying the sentiment and context of what is being spoken. This paper aims to present a new methodology to create an emotionally expressive virtual AI that is capable of understanding the sentiment of the conversation and displaying emotions during conversations. To achieve this the system uses a generative chatbot and combines it with a 3D talking head that is animated parametrically. This work could be beneficial to virtual assistants and help facilitate more lifelike interactions, holding significance in environments that require the user to feel more comfortable with their interactions. The complexities lie in developing a domain specific chat-bot that will not only provide valuable replies but also recognize and display appropriate facial expressions while communicating.