The major objective of the proposed research is to identify some of the talker and listener characteristics that contribute to variation in speech intelligibility. The study consists of an extensive examination of the production and perception of """"""""clear speech,"""""""" a distinct speaking style that talkers adopt when the listener has speech perception difficulties due to a hearing loss, background noise, or a different native language. To the extent that clear speech is more intelligible than """"""""conversational speech,"""""""" an acoustic-phonetic comparison of these two speaking styles provides unique information about factors that affect speech intelligibility. The proposed studies will test the hypothesis that naturally produced clear speech reflects an interaction of universal, auditory-perceptual factors, which serve to enhance the overall acoustic salience of the speech signal such that it is more resistant to the adverse effects of environmental noise or listener-related perceptual deficits, and 1anguage-specific structural factors, which serve t o e enhance t he realization of phonologically important contrasts. Two important predictions of this hypothesis are (a) clear speech production will show predictable and systematic similarities and differences across languages, and (b) the intelligibility benefit of naturally produced clear speech will be greater for listeners with well-entrenched knowledge of the sound structure of the target language than for listeners with limited experience with the sound structure of the target language. In order to test these predictions two projects are proposed. Project 1 will investigate variability in clear speech production by comparing the conversational-to-clear speech transformations in English and Spanish. Project 2 will investigate variability in clear speech perception by comparing the clear speech intelligibility benefit across listeners that vary with respect to their experience with the sound structure of the target language (native vs. non-native listeners, adults vs. children).
Showing the most recent 10 out of 36 publications
