Making Speech Synthesis More Accessible to Older People

Maria Wolters, Pauline Campbell, Christine DePlacido, Amy Liddell, David Owens

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this paper, we report on an experiment that tested users’ ability to understand the content of spoken auditory reminders. Users heard meeting reminders and medication reminders spoken in both a natural and a synthetic voice. Our results show that older users can understand synthetic speech as well as younger users provided that the prompt texts are well-designed, using familiar words and contextual cues. As soon as unfamiliar and complex words are introduced, users’ hearing affects how well they can understand the synthetic voice, even if their hearing would pass common screening tests for speech synthesis experiments. Although hearing thresholds correlate best with users’ performance, central auditory processing may also influence performance, especially when complex errors are made.
Original languageEnglish
Title of host publicationSixth ISCA Workshop on Speech Synthesis
Publication statusPublished - Aug 2007
EventSixth ISCA Workshop on Speech Synthesis - Bonn, Germany
Duration: 22 Aug 200724 Aug 2007

Workshop

WorkshopSixth ISCA Workshop on Speech Synthesis
CountryGermany
CityBonn
Period22/08/0724/08/07

Fingerprint Dive into the research topics of 'Making Speech Synthesis More Accessible to Older People'. Together they form a unique fingerprint.

Cite this