Method Of Visual Communication On Speech Translating System Based On Avatar

摘要

PURPOSE: A method for video communication of an avatar based TTS system is provided to synchronize the shape of lips of an avatar with a synchronized voice in order to control the look and movement of the avatar in response to the contents of speech, and thereby perform the video communication . CONSTITUTION: An avatar model producer determines whether it performs an operation for making an online photograph or selects an existing avatar model(301). The online photograph is made(302), and a reference point of the avatar model is illustrated(303) if the online photograph is needed. If the existing avatar model is selected, the data of the selected avatar model is transmitted to an avatar server(305). A voice recognition module recognizes the voice inputted from an outside and thereby generates a character string, and transmits it to a language translation module(306). The language translation module translates the generated character string and transmits it to a voice synthesizing module(307). The voice synthesizing module extracts the information related to the movement from the translated character string.