28 Years of Vowels: Effects of talker age on pronunciation

Susanne Gahl
UC Berkeley, Department of Linguistics

Teenagers sound different from centenarians. One would be surprised to learn otherwise, and plenty of research has documented the effects of vocal aging. But do thirty-year olds sound different from forty-year olds? Relatively little is known about the nature of such differences, and even less about their relationship to adult vocabulary development or learning generally. One reason for this gap in knowledge is an assumption commonly made in psycholinguistic research, which is that the language processing system remains fairly stable from the end of puberty until about age 70. Accordingly, literature on age-related changes in speech has primarily focused on comparing “young and middle age” adults (18–50 years of age) to “elderly” adults (75+ years). And yet, listeners are able to guess talker age far more accurately than a binary distinction would imply. Clearly, acoustic characteristics of speech must change continually and gradually throughout adulthood.

In this talk, I present an analysis of speech samples from a documentary film series (the ‘Up’ series directed by Michael Apted) following eleven young-to-middle-aged individuals over a period of 28 years. I show that duration and resonant frequencies of vowels change in a systematic fashion as talkers move from young to middle-age adulthood. I discuss the implications of the nature and direction of that change for models of pronunciation variation and language production.

This talk contains joint research with Harald Baayen.

If you would like to meet with the speaker, please contact Bistra Andreeva.