Weighting of Auditory Feedback Across the English Vowel Space.

Abstract

Auditory feedback in the headphones of talkers was manipulated in the F1 dimension using a real-time vowel formant filtering system. Minimum formant shifts required to elicit a response and the amount of compensation were measured for vowels across the English vowel space. The largest response in production of F1 was observed for the vowel /ε/ and smaller or non-significant changes were found for point vowels. In general, changes in production were of a compensatory nature that reduced the error in the auditory feedback.