Remez R E
Department of Psychology, Barnard College, New York, NY 10027-6598, USA.
J Acoust Soc Am. 1996 Mar;99(3):1695-98. doi: 10.1121/1.414693.
Some influential accounts of speech perception have asserted that the goal of perception is to recover the articulatory gestures that create the acoustic signal, while others have proposed that speech perception proceeds by a method of acoustic categorization of signal elements. These accounts have been frustrated by difficulties in identifying a set of primitive articulatory constituents underlying speech production, and a set of primitive acoustic-auditory elements underlying speech perception. An argument by Lindblom favors an account of production and perception based on the auditory form of speech and its cognitive elaboration, rejecting the aim of defining a set of articulatory primitives by appealing to theoretical principle, while recognizing the empirical difficulty of identifying a set of acoustic or auditory primitives. An examination of this thesis found opportunities to defend some of its conclusions with independent evidence, but favors a characterization of the constituents of speech perception as linguistic rather than as articulatory or acoustic.
一些有影响力的言语感知理论认为，感知的目标是还原产生声学信号的发音动作，而另一些理论则提出，言语感知是通过对信号元素进行声学分类的方法来进行的。由于难以确定构成言语产生基础的一组原始发音成分，以及构成言语感知基础的一组原始声学 - 听觉元素，这些理论一直受到阻碍。林德布洛姆的一个观点支持基于言语的听觉形式及其认知细化来解释言语产生和感知，反对通过诉诸理论原则来定义一组发音原语的目标，同时也认识到确定一组声学或听觉原语在经验上的困难。对这一论点的审视发现，有机会用独立证据来支持其一些结论，但更倾向于将言语感知的成分描述为语言学的，而非发音或声学的。
