Linguistic and Phonetic Approach to Fluency Disorders

本研究试图阐明日语、英语和韩语中不流利现象背后的语音和音系模式及环境。这些模式和环境也许能给我们提供一个寻找言语不流利因素的途径。它们还可以为我们提供语言细节的言语产生过程的线索。为了做到这一点，分析的对象集中在重复风格的不流利。在正常的连贯言语中，口吃者和非口吃者都经常或有时会产生这种重复。因此，把它们提取出来作为言语样本并不困难。收集好的言语不流利样本是研究的第一阶段。在韩国，约3500个口吃样本和背景样本是在韩国口吃自助小组会议上的演讲中收集的。下一个收集的样本是汉语不流利。但是SARS阻止了中国的样本，所以下一步是分析收集的韩国样本 关于我们 包括来自电视脱口秀节目的非口吃者的不流利样本。而不是中国样品。通过8小时的印尼电视访谈节目收集了印尼非口吃者的不流利样本和背景样本。但是到2004年底还没有找到从印尼语中提取口吃者样本的方法。因此，只有口吃者和非口吃者产生的韩语不流利语可以与其他语言、日语和英语进行分析和对比。它被组织成流畅性障碍的跨语言研究。分析的总样本约为19，000个，包括背景样本。本研究揭示了成人言语不流利的三个显著特征：（1）大多数不流利出现在词首;（2）重复单位反映了音系单位（CVC音节的一个组成部分：起始、莫拉或音节本身）。但每种语言的重复次数都相当多，将CVC音节分解为CV和V，且口吃者的C单元重复次数多于非口吃者。（iii）从语音转换缺陷的角度来看，统计分析表明，在日语、英语和韩语中，口吃者和非口吃者之间出现不流利的环境有显著的不同。塞音（塞音）是口吃者言语不流利的触发器。另一方面，鼻音使非口吃者的语言不流利。这一结果表明，口吃者和非口吃者的不流利表达具有不同的背景，即不同的不流利表达产生过程，尽管这两种不流利表达的形式是相同的。少

This research attempts to illuminate the phonetic and phonological patterns and environments underlying disfluency in Japanese, English and Korean. The patterns and the environments may be able to give us a way to find the factor of speech disfluency. They may also show us a hint on the process of speech production of language specifics. To do this, the object for analysis focuses on a repetition style of disfluency. In the normal connected speech, both stuterers and nonstutterers often or sometimes produce this kind of repetition. So it's not difficult to pick them up as speech samples.Collecting good samples of speech disfluency was the first stage of the research. In Korea about 3500 stuttering samples and the background samples were picked up in the speeches at the meeting of the Korean stuttering self-help group. The next collecting samples would have been Chinese disfluency. But Sars prevented from getting Chinese samples.So the next stage was to analyze collected Korean samples … More including nonstutterers' disfluency samples from TV talk show programs. Instead of Chinese samples. Indonesian nonstutterers' disfluency samples and the background samples were collected through 8-hours Indonesian TV talk show programs. But the way to get stutters' samples in Indonesian could not be found by the end of 2004.Therefore, only Korean disfluency produced by stutterers and nonstutterers can be analyzed and contrasted with the other languages, Japanese and English. It is organized into The Cross-Linguistic Study on Fluency Disorders. Analyzed total samples are about 19,000 including background samples. Japanese are English samples are from previous studies by the same author (Ujihira 2000, 2003).This analysis has revealed three salient characteristics of disfluency in adult speech : (i)Most of the disfluency appears at the word initial, (ii)The unit of repetition reflects the phonological unit (a constituent of CVC syllable : onset, mora, or syllable itself)). But considerable number of repetitions of each language breaks CVC syllables into CV and V. And the C-unit repetition in stuttterers' samples is more than that in nonstutterers'. (iii)From the viewpoint of a phonetic transition defect, the statistic analysis has revealed that the environments where disfluency occurs are remarkably different between stutterers and nonstutterers in Japanese, English, and Korean. The stops (Plosives) are trigger for disfluency in stutterers' speech. On the other hand the nasals draws disfluency in nonstutterers' speech. This result suggests that the stutterers' disfluency and the nonnstutterers' kind have different backgrounds, in other words different processes for production of disfluency, though both of the styles of disfluency are the same form. Less