Supra-Segmental Changes in Speech Production as a Result of Spectral Feedback Degradation: Comparison with Lombard Speech.
Elizabeth D Casserly, Yeling Wang, Nicholas Celestin and 2 others
PMID 28653556WHAT IT FOUND
Normal-hearing speakers slowed down, raised pitch and loudness, then became quieter and lower-pitched when hearing simulated cochlear implant feedback.
Listeners preferred their baseline speech and recognized feedback-altered speech more slowly.
Key findings
01With CI-simulated feedback, sentence duration increased from 1.648 s at baseline to 1.854 s, then decreased to 1.709 s.
02Average pitch and average loudness in experimental speakers rose initially and then fell below baseline: pitch went 189.6 to 193.7 to 184.1 Hz, and loudness went 66.2 to 67.0 to 64.63 dB.
03Listeners judged baseline words from experimental talkers easier to understand 55.45% of the time, and recognized words produced with feedback more slowly (2.09 s) than baseline words (2.02 s).
STILL TO COME
How it was doneWhat they foundWhat it means for SLPs
Read the rest of this summary
You get three full summaries a month, free, and we do not ask for a card. Search, the TL;DRs and your library stay unlimited either way.
What it does not show
Speech production data came from 12 adults, with 9 experimental and 3 controls, and the controls were all female. Participants were normal-hearing adults hearing simulated cochlear implant feedback, not people who use cochlear implants. Loudness was recorded with a stand microphone, so changes in distance from the microphone could affect amplitude measures. The experimental talkers were more intelligible than controls even at baseline, making talker-group comparisons harder to interpret. Pitch variability changes after 15 minutes were hard to interpret because baseline and later conditions did not show clear contrasts. Intelligibility was tested with isolated words, not sentences, because sentence scoring can be influenced by context.
The easy way to misread this
Do not conclude that CI-simulated feedback creates clearer speech. Listeners preferred baseline speech and recognized feedback-altered speech more slowly, and the study used normal-hearing speakers hearing simulated feedback, not clinical CI users.