可视化协同发音合成研究综述

A Survey of the Synthesis of Visual Co-Articulation

可视语音应用于教学、通信、电子商务等领域,可进一步提高人机交互的友好性和方便性,近年来得到广泛关注。可视化协同发音合成是研究可视语音的重要环节之一,阐述可视化协同发音的基本概念,介绍有代表性的描述方法,并对基于图像和基于模型方法这两种重要的合成方法的研究现状进行评述。总结两种方法的优缺点,并展望其发展方向。

Visual speech, used in the fields of teaching, communication, e-commerce etc. to further improve the friendliness and ease of human-computer interaction, has aroused wide concern in recent years. As one of the important parts of the research on visual speech, introduces the concept of visual co-articulation as well as some representative descriptive methods and a review of the two important methods based on image and model respectively. Summarizes the advantages and disadvantages of two methods and prospects the way of its development.