Summers W V, Johnson K, Pisoni D B, Bernacki R H
Department of Psychology, Indiana University, Bloomington 47405.
J Acoust Soc Am. 1989 Nov;86(5):1717-21. doi: 10.1121/1.398602.
The authors respond to Fitch's comments [H. Fitch, J. Acoust. Soc. Am. 86, 2017-2019 (1989)] on an earlier paper. New analyses are presented to address the question of whether F1 differences observed in the original report are an artifact of linear predictive coding (LPC) analysis techniques. Contrary to Fitch's claims, the results suggest that the F1 differences originally reported are, in fact, due to changes in vocal tract resonance characteristics. It is concluded that there are important acoustic-phonetic differences in speech when talkers speak in noise. These differences reflect changes in both glottal and supraglottal events that are designed to maintain speech intelligibility under adverse conditions.
作者们回应了菲奇对其早期一篇论文的评论[H. 菲奇，《美国声学学会杂志》86, 2017 - 2019 (1989)]。文中给出了新的分析，以探讨在原始报告中观察到的F1差异是否是线性预测编码（LPC）分析技术的人为产物这一问题。与菲奇的观点相反，结果表明最初报告的F1差异实际上是由于声道共振特征的变化所致。得出的结论是，当说话者在噪声环境中讲话时，语音存在重要的声学语音差异。这些差异反映了声门和声门上事件的变化，这些变化旨在在不利条件下维持语音可懂度。
