基于三维可视语音库的发音康复方法  

Pronunciation rehabilitation methods based on 3D visualization speech database

作　　者：赵剑[1] 王丽荣[2] 

机构地区：[1]长春大学计算机科学技术学院,吉林长春130022 [2]长春大学科研处,吉林长春130022

出　　处：《长春大学学报》2010年第2期75-77,共3页Journal of Changchun University

基　　金：吉林省科技发展计划项目(20082111);吉林省教育厅"十一五"科学技术研究项目(2009232;2009446;2009451);长春市国际科技合作计划项目(08GH07)

摘　　要：将语音合成领域中研究的可视语音技术应用于康复工程之中,设计了一个基于三维视觉的语音库系统。提出了基于三维语音库的语音辅助系统应用于聋儿语言发音训练的新方法,通过直观的视觉反馈对聋而不哑的听障儿童进行发音训练,从而改善他们的发音质量。提出该方法的框架结构,系统能够培养聋儿语感及自主学习语言的意识,使其具有一定的说话能力,为其今后的知识学习和参与社会活动奠定基础。A speech database system based on 3D visualization is designed by applying the visualization speech technology of speech synthesis into rehabilitation project. A new method that speech aid system based on 3D visualization speech database should be used in language pronunciation training for deaf children is proposed. The method is used to train hearing-impaired children to pronounce through intuitive visual feedback, so as to improve their quality of pronunciation. The framework can foster language sense of deaf chil- dren and the consciousness of their autonomous learning language. The system will make deaf children capable in speaking and will lay a foundation for their learning and participating in social activities.

关 键 词：语音合成 三维语音库 可视语音 聋儿康复 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]