ABSTRACT
In recent years, a theory of acoustic invariance has been elaborated which attempts to account for the speech processing mechanism in man, as well as provide a theoretical framework to characterize the nature of the phonological systems of natural language (of. Stevens & Blumstein, 1979, 1981; Blumstein & Stevens, 1981). This theory has been guided by two claims. The first is that there is acoustic invariance in the speech signal corresponding to the phonetic features of natural language. That is, it is hypothesized that the speech signal is highly structured in that it contains invariant acoustic patterns for phonetic features, and these patterns remain invariant across speakers, phonetic contexts, and languages. The second claim is that the perceptual system is sensitive to these invariant properties. That is, it is hypothesized that the perceptual system can use these invariant patterns to provide the phonetic framework for natural language, and to process the sounds of speech in ongoing perception.
