Tiffany W R
J Speech Hear Res. 1980 Dec;23(4):894-908. doi: 10.1044/jshr.2304.894.
Paragraphs with controlled phonetic structures were matched to similarly structured diadochokinetic (Maximum Repetition Rate) tasks in an effort to devise a more valid measurement for (1) assessing possible relationships between diadochokinesis and speech rate, and (2) evaluating the effects on articulation rates of such structural variables as number of consonants in a syllable, and alternating versus simple syllable repetitions. Highly stable results were obtained, suggesting the possibility of a sharp neurophysiological or biomechanical barrier which varies markedly among presumably normal speakers. Maximum repetition rates were poor predictors of normal reading rate performance. On the other hand, normal reading rates were found to be approximately the same as the maximum repetition rates--about 13.5 phones per second. The inference is that normal speech is not, as commonly supposed, obviously slower than maximum rates of syllable articulation, for equivalent syllables. The major source of variation in syllable rate measures was simply the number of phones in a syllable. The effects of articulatory place and manner appeared relatively trivial by comparison.
具有可控语音结构的段落与结构相似的连续交替运动速率(最大重复率)任务相匹配,目的是设计一种更有效的测量方法,用于:(1)评估连续交替运动速率与语速之间可能存在的关系;(2)评估诸如音节中辅音数量以及交替音节重复与简单音节重复等结构变量对发音速率的影响。获得了高度稳定的结果,这表明可能存在一个明显的神经生理或生物力学障碍,在假定正常的说话者中,这个障碍存在显著差异。最大重复率并不能很好地预测正常阅读速率表现。另一方面,发现正常阅读速率与最大重复率大致相同——约为每秒13.5个音素。由此推断,正常言语并不像通常认为的那样,对于同等音节,明显慢于音节发音的最大速率。音节速率测量中变化的主要来源仅仅是音节中的音素数量。相比之下,发音部位和方式的影响显得相对微不足道。