权益分类	功能权益	普通用户	{{item.name}}会员
{{category.name}}	{{benefitItem.name}}

Research on the development and the make of the assistive technology tools to promote communication ability for person with intellectual disabilities

提升智障人士沟通能力辅助技术工具的开发与制作研究

基本信息

批准号：
16200048
负责人：
NAKABAYASHI Toshitaka
金额：
$ 26.79万
依托单位：
Kobe University
依托单位国家：
日本
项目类别：
Grant-in-Aid for Scientific Research (A)
财政年份：
2004
资助国家：
日本
起止时间：
2004 至 2006
项目状态：
已结题

来源：
https://kaken.nii.ac.jp/en/grant/KAKENHI-PROJECT-16200048/
关键词：
Intellectual Disability VOCA Communication Assistive tool Speech Recognition 知的障害者

项目摘要

1. Autistic disorder is a neurobiological disorder of development for one's lifetime. Children with autistic disorder have difficulty communicating and interacting with others. We developed and tested a suit of five PDA application software "Picture Aid" for Children with intellectual disability (include autistic disorder) need extensive or pervasive supports. (1) aid for ordinal procedure, (2) timer,. (3) picture card with voice, (4)words slot machine game with picture, and (5) puzzle. We designed our aid for ordinal procedure with many customizable pictures. Customization is done on a PC. After customization, user transmit new application software to PDA.2. Recently, the accuracy of speaker-independent speech recognition has been remarkably improved by use of stochastic modeling of speech. However, the use of those acoustic models causes degradation of speech recognition for a person with different speech style (e.g., articulation disorders). We had tried to build the acoustic model for a person with articulation disorders (a person with articulation disorders by athetoid type of cerebral palsy). The articulation of the first utterance tended to become unstable due to strain of a muscle and that caused degradation of speech recognition, where MFCC (Mel Frequency Cepstral Coefficients) iwas used as speech features. Therefore we proposed a robust feature extraction method based on PCA (Principal Component Analysis) instead of MFCC. Its effectiveness was confirmed by word recognition experiments.

1.自闭症是一种终生发育的神经生物学障碍。患有自闭症的孩子很难与他人沟通和互动。我们开发并测试了一套五款PDA应用软件“图片辅助器”，用于需要广泛或普遍支持的智力残疾儿童（包括自闭症儿童）。(1)用于顺序程序的辅助装置，（2）定时器，. (3)带语音的图片卡，（4）带图片的文字老虎机游戏，和（5）拼图。我们设计了我们的援助顺序程序与许多可定制的图片。定制是在PC上完成的。定制后，用户将新的应用软件传输到PDA.近年来，通过对语音进行随机建模，非特定人语音识别的准确率得到了显著提高。然而，这些声学模型的使用导致具有不同语音风格的人的语音识别的降级（例如，发音障碍）。我们曾试图为一个有发音障碍的人（一个有手足徐动型脑瘫的发音障碍的人）建立声学模型。第一个发音的清晰度往往会由于肌肉的应变而变得不稳定，这会导致语音识别的退化，其中MFCC（Mel频率倒谱系数）被用作语音特征。因此，我们提出了一种鲁棒的特征提取方法，基于主成分分析（PCA），而不是MFCC。其有效性得到了单词识别实验的证实。