Abstract
Studies of deaf speech frequently cite faulty timing as a major cause of poor intelligibility. Yet modifying the timing of normal speech seems to have so little effect on its intelligibility that formal quantification is not worthwhile. Usually, however, the person who modifies the timing also knows what the utterance said, and therefore is not forced to rely on what he or she hears to establish its prosodic pattern. When formal tests are run, sentences that appeared perfectly acceptable to the experimenter suddenly become unintelligible to the subjects. An experiment will be reported comparing the intelligibility of monotone speech synthesized by rule under four conditions: (1) segment durations assigned by rule, (2) equal duration for all segments, (3) equal duration for all syllables, and (4) equal duration for all metric feet.
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have