Technical Reports (CIS)

Document Type

Technical Report

Date of this Version

November 1990


University of Pennsylvania Department of Computer and Information Science Technical Report No. MS-CIS-90-88.


Our goal is to build a system of 3-D animation of facial expressions of emotion correlated with the intonation of the voice. Up till now, the existing systems did not take into account the link between these two features. Many linguists and psychologists have noted the importance of spoken intonation for conveying different emotions associated with speakers' messages. Moreover, some psychologists have found some universal facial expressions linked to emotions and attitudes. We will look at the rules that control these relations (intonation/emotions and facial expressions/emotions) as well as the coordination of these various modes of expressions. Given an utterance, we consider how the message (what is new/old information in the given context) transmitted through the choice of accents and their placement, are conveyed through the face. The facial model integrates the action of each muscle or group of muscles as well as the propagation of the muscles' movement. It is also adapted to the FACS notation (Facial Action Coding System) created by P. Ekman and W. Friesen to describe facial expressions. Our first step will be to enumerate and to differentiate facial movements linked to emotions from the ones linked to conversation. Then, we will examine what the rules are that drive them and how their different actions interact.


facial animation, emotion, intonation, coarticulation, conversational signals



Date Posted: 05 August 2007