Abstract

The current study examined the change in articulatory patterns when speakers were asked to increase and decrease their speaking rate by matching metered and rhythmic audio recordings and by matching metered and rhythmic audiovisual recordings. There were 10 participants, five male and five female, ranging in age from 20 to 36 with a mean age of 25. Participants spoke the sentence "Don't fight or pout over a toy car"under rhythmic, metered, fast and slow conditions and in response to audio only or audiovisual models, resulting in eight speaking conditions: audio metered fast, audiovisual metered fast, audio metered slow, audiovisual metered slow, audio rhythmic fast, audiovisual rhythmic fast, audio rhythmic slow, and audiovisual rhythmic slow. Each participant had five sensors glued to their tongue, teeth and lips and articulatory movements were recorded with an NDI Wave electromagnetic articulograph. 10 tokens of the target utterance were analyzed for duration and Spatiotemporal Index (STI). STI was then computed for the vertical movements of the tongue, jaw and lower lip, as well as lip aperture in order to measure variability of speech movements over multiple sentence repetitions. Stroke metrics based on the speed history of the articulators were also computed in order to reveal average kinematic features of articulatory gestures, or the individual. movement strokes that occurred between successive speed minima in running speech. Statistical analysis revealed that STI measures did not change significantly in response to the different rate conditions. This study demonstrated that in neurotypical individuals, articulatory patterns including stroke count, onset speed, peak speed and hull area changed significantly in faster or slower speech. Additionally, the results revealed that both metered and rhythmic cues and both audio and audiovisual cues are effective in decreasing and increasing speaking rate without significantly impacting the STI (i.e., consistency) of articulatory movements. Therefore, it may be that a speaker's efforts to match the audio and audiovisual cues in real-time more significantly affected articulation patterns than whether cues were rhythmic, metered, audio or audiovisual.

Degree

College and Department

David O. McKay School of Education; Communication Disorders

Rights

https://lib.byu.edu/about/copyright/