Perception of speaker sex and vowel recognition at very short durations

Abstract

A man or woman saying the same vowel do so with very different voices. The auditory system solves the problem of extracting what the man or woman said despite substantial differences in the acoustic properties of the carrier voice. Much of this acoustic variation is due to differences in the underlying anatomical mechanisms for producing speech. If the auditory system knew the sex of the speaker then it could correct for speaker-sex related acoustic differences thus facilitating vowel recognition. We measured the minimal stimulus duration necessary to accurately discriminate whether a brief vowel segment was spoken by a man or woman, and to accurately recognize what vowel was spoken. Results show that reliable vowel recognition precedes reliable speaker sex discrimination. Furthermore, the pattern of performance across experiments where voice pitch and resonance information were systematically varied, is markedly different depending on whether the task is speaker-sex discrimination or vowel recognition. These findings suggest that knowledge of speaker sex has little impact upon vowel recognition at very short stimulus durations.