Andrews G, Howie P M, Dozsa M, Guitar B E
J Speech Hear Res. 1982 Jun;25(2):208-16.
Speech samples were collected from three adult male stutterers under six baseline conditions and 15 conditions believed to increase fluency. After moments of stuttering and filled pauses had been deleted from the samples, a speech pause analysis technique developed by Goldman-Eisler was used to measure the following speech pattern characteristics: mean phonation duration (i.e., the duration between pauses), pause proportion, articulation rate, fluent speech rate, mean sentence length, and percentage of syllables stuttered. Greater than 50% reduction in stuttering occurred in all but three conditions (Speak and Write, Relaxed, Alone with cards). Greater than 90% reduction occurred under Prolonged/DAF speech, Singing, Chorus reading, Shadowing, Slowing, Syllable-timed speech, and Response contingent stimulation. The data were examined for evidence of speech pattern characteristic changes which were associated with reduced stuttering. Lengthened mean phonation duration occurred consistently under four conditions: Chorus reading, Shadowing, Singing, and Prolonged/DAF. Slowed speech (lower overall rate, lower articulation rate, or increased pause proportion) occurred consistently in seven conditions: Prolonged/DAF, Slowing, Syllable-timed speech, Arm swing, Speak and write, Relaxed, and Singing. Only in Prolonged/DAF speech did lengthened phonation duration occur in conjunction with slowed speech. The results of this exploratory study suggest that stuttering may be reduced under different conditions by means of different strategies. Lengthened phonation and slowing were the predominant strategies used in those conditions investigated in this study. The results are consistent with those of effective treatment techniques. Theoretical accounts of the association between change in fluency and change speech pattern characteristics are discussed.
从三名成年男性口吃者身上收集了语音样本，这些样本处于六种基线条件下以及被认为能提高流畅性的15种条件下。在从样本中删除口吃时刻和填充停顿后，使用戈德曼 - 艾斯勒开发的语音停顿分析技术来测量以下语音模式特征：平均发声持续时间（即停顿之间的持续时间）、停顿比例、发音速率、流畅言语速率、平均句子长度以及口吃音节的百分比。除了三种条件（边说边写、放松、独自看卡片）外，在所有其他条件下口吃减少幅度均超过50%。在延长/延迟听觉反馈言语、唱歌、齐声朗读、跟读、放慢语速、音节定时言语和反应性刺激条件下，口吃减少幅度超过90%。检查数据以寻找与口吃减少相关的语音模式特征变化的证据。在四种条件下（齐声朗读、跟读、唱歌和延长/延迟听觉反馈），平均发声持续时间持续延长。在七种条件下（延长/延迟听觉反馈、放慢语速、音节定时言语、摆动手臂、边说边写、放松和唱歌），言语放慢（总体速率较低、发音速率较低或停顿比例增加）持续出现。只有在延长/延迟听觉反馈言语中，发声持续时间延长与言语放慢同时出现。这项探索性研究的结果表明，在不同条件下可以通过不同策略减少口吃。延长发声和放慢语速是本研究调查的那些条件中使用的主要策略。结果与有效治疗技术的结果一致。讨论了流畅性变化与语音模式特征变化之间关联的理论解释。
