基于剧本驱动的双主体可视语音研究  

Research on Text-to-visual Speech (TTVS) Based on Double-avatar of Script-driven

作　　者：姚军[1] 杨志晓[1] 

机构地区：[1]河南工业大学信息科学与工程学院,郑州450001

出　　处：《计算机与数字工程》2009年第2期76-78,148,共4页Computer & Digital Engineering

摘　　要：提出了基于剧本驱动的双主体可视语音的概念。传统的文本信息很难表达文本所包含的全部信息。将普通的文本转换为剧本,同时将文字语音化,把隐含在文字中的非语言信息,通过虚拟机器人以适当的肢体语言表达出来。虚拟机器人是剧本信息提供者的主体。通过提供者的主体,以剧本为驱动能更好的传达隐含在文本中的非语言信息,提高文本传输的效率。A concept of TTVS based on double-avatar of script-driven is proposed. Traditional text can not express all information embedded in contexts. The text will be converted into ordinary script and text is changed into speeches, nonverbal information which embedded in contexts is expressed with body languages through virtual robot. The virtual robot is called the avatar of the person who gives script. An avatar based on script-driven can expressively exchange not only nonverbal information of the text, but also improving the efficiency of the text.

关 键 词：剧本 化身 虚拟机器人 非语言信息 可视语音 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]