Um, so, like, do speech disfluencies matter? A parametric evaluation of filler sounds and words

Abstract

This study evaluated how speech disfluencies affect perceived speaker effectiveness. Speeches with filler sounds and filler words at different rates of disfluencies (i.e., 0, 2, 5, and 12 per minute) were created and evaluated by a crowdsourcing service for survey-based research for the speaker's public speaking performance. Increased disfluencies, particularly filler sounds, significantly affected perceptions across most categories, notably at higher rates of filler sounds (i.e., 12 per minute). A low, but nonzero, rate of disfluencies (5 per minute) did not adversely affect perceived effectiveness. These findings suggest that although reducing filler sounds is crucial for optimizing perceived speaking effectiveness, a rate of five or fewer disfluencies per minute may be acceptable.