No open-access location is known for this record yet.
No open-access location is known for this record yet.
Understanding how orofacial muscle activity and movement covary across changes in speech rate and intensity has implications for the neural control of speech production and the use of clinical procedures that manipulate speech prosody. The present study involved a correlation analysis relating average lower-lip and jaw-muscle activity to lip and jaw movement distance, speed, and duration. Recordings were obtained on orofacial movement, muscle activity, and the acoustic signal in 3 normal speakers as they repeated a simple test utterance with targeted speech rates varying from 60% to 160% of their habitual rate and at targeted vocal intensities of -6 dB and +6 dB relative to their habitual intensity. Surface electromyographic (EMG) recordings were obtained with electrodes positioned to sample primarily the mentalis, depressor labii inferior, anterior belly of the digastric, and masseter muscles. Two-dimensional displacements of the lower lip and jaw in the midsagittal plane were recorded with an electromagnetic system. All participants produced linear changes in percent utterance duration relative to the auditory targets for speech rate variation. Intensity variations ranged from -10 dB to +8 dB. Average EMG levels for all 4 muscles were well correlated with specific parameters of movement. Across the intensity conditions, EMG level was positively correlated with movement speed and distance in all participants. Across the rate conditions, EMG level was negatively correlated with movement duration in all participants, while greater interparticipant variability was noted for correlations relating EMG to speed and distance. For intensity control, it is suggested that converging neural input to orofacial motoneurons varies monotonically with movement distance and speed. In contrast, rate control appears to be more strongly related to the temporal characteristics of neural input than activation level.
Loading…
Stored only when both papers are in the corpus — 54 citations exist upstream.
Papers that cite the same references — the relational fallback while embeddings are still backfilling.
Imported from OpenAlex (CC0) — nothing here is generated.