5 - Theories of Speech Perception

Abstract

This chapter discusses the theories of speech perception. The process of speech perception requires an analysis and decoding of the acoustic information available in the sound pattern. The auditory receptor system receives a sound pattern that can be described by fluctuations in sound pressure over time. The feature detection process determines the presence or absence of the acoustic features critical to speech perception. In feature detection, the acoustic information can be analyzed as it arrives at the auditory receptor system. The set of acoustic features is transferred and held in a second structural component, preperceptual auditory storage, where it remains for about 250 msec. Information in the signal must be stored as the various acoustic features are dispersed throughout the sound pattern corresponding to one perceptual unit and will therefore be detected at different points in time, and the primary recognition process will take time. The primary recognition—perception—process transforms the preperceptual pattern of acoustic features into a synthesized percept in a third structural component, synthesized auditory memory. The outcome of any recognition process is the report that one of a possible set of alternatives is present. Primary recognition occurs when the acoustic features can uniquely determine a perceptual unit of speech.