AUTOMATED GENERATION OF ANIMATED WEATHER REPORTS WITH A VIRTUAL WEATHERMAN

摘要

This paper proposes automatic generation of animated weather reports, where a virtual human performs as a weatherman, as an example of automated TV-like program production. An input to our prototype system is a set of (location, time, weather)-type weather forecast data. The system then summarizes the data and makes a weather report scenario in the form of speech texts with embedded commands, most of which relate to the virtual weatherman's body language. From this command-embedded speech text, our virtual human presenter system makes the virtual human agent act as a weatherman with the weather charts in real-time 3D animation synchronized with speech outputs. When the system runs on a Web server, on-demand speech-text generation enables us to watch personalized and up-to-the-minute weather reports on the Web.