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

The Computer Based Acoustic Analysis of Speech Sounds for Assessing Motor Speech Disorders.

基于计算机的语音声学分析，用于评估运动言语障碍。

基本信息

批准号：
10838031
负责人：
SEKI Hideaki
金额：
$ 1.54万
依托单位：
Chiba Institute of Technology
依托单位国家：
日本
项目类别：
Grant-in-Aid for Scientific Research (C)
财政年份：
1998
资助国家：
日本
起止时间：
1998 至 1999
项目状态：
已结题

来源：
https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-10838031/
关键词：
Motor speech disorders Acoustic analysis Auditory perceptual analysis Power change Spectrum change speed Duration of the frame sentences Fundamental frequency DP matching

项目摘要

The computer based acoustic analysis of speech sounds for assessing motor speech disorders was proposed. The relationship between the results of acoustic analysis and that of auditory perceptual analysis judged by speech therapists were also discussed.The speech samples of this sturdy were consisted of frame sentences "I I apapa desu" and "I I akaka desu" which contained consonants /p/ and /k/. And the free utterances were also obtained from the patients. The speech materials were obtained from 46 patients with motor speech disorders and 66 normal people as controls. The parameters for acoustic analysis were the power change of speech sound, the spectrum change speed of speech sound, the ratio of fundamental frequency change, the duration of the frame sentences and the values by DP matching method. The items of auditory perceptual analysis were the rhythm of speech, the rate of speech, the imprecise consonants, the speech intelligibility and bizarreness, the severity of motor speech disorders (sum of the scale values of bizarreness and intelligibility).There were high correlation between the imprecise consonants and the power change of speech sound, the bizarreness and the duration of the frame sentences, the speech intelligibility and the ratio of fundamental frequency change, and between the severity of motor speech disorders and the values by DP matching method. These results suggested that the proposed acoustic analysis of speech sounds for assessing motor speech disorders may help inexperienced speech therapists to evaluate the speech of patients with motor speech disorders objectively.

提出了一种基于计算机的语音声学分析方法来评估运动语言障碍。讨论了语音分析结果与言语治疗师判断的听觉知觉分析结果之间的关系。该实验的语音样本由包含辅音/p/和/k/的框架句“I I apapa desu”和“I I akaka desu”组成。并获得患者的自由话语。言语材料取自46例运动言语障碍患者和66例正常人作为对照。声学分析的参数是语音的功率变化、语音的频谱变化速度、基频变化比、框架句子的持续时间和DP匹配方法的值。听觉知觉分析项目为言语节奏、语速、不精确辅音、言语可理解性和奇异性、运动言语障碍严重程度（奇异性和可理解性量表值之和）。不精确辅音与语音的力度变化、框架句的奇异性与持续时间、语音可理解性与基频变化比、运动言语障碍的严重程度与DP匹配值有较高的相关性。这些结果表明，将语音的声学分析用于评估运动语言障碍可以帮助缺乏经验的语言治疗师客观地评估运动语言障碍患者的语言。