14thAnnual Conference of the International Speech Communication Association

Lyon, France
August 25-29, 2013

Assessing the Utility of Judgments of Children's Speech Production Made by Untrained Listeners in Uncontrolled Listening Environments

Benjamin Munson

University of Minnesota, USA

Speech perception experiments conducted on the internet have numerous benefits over traditional laboratory studies. For example, they have the potential benefit of including many more participants than can be recruited in traditional laboratory-based studies, including members of special hard-to-recruit groups. In the study of phonological development, they have the potential for generating judgments of children's speech production accuracy that reflect the consensus of the child's speech community, rather than the judgments of a small group of phonetically trained individuals. Internet-based studies have the potential disadvantage of being conducted in less controlled listening environments with more variable equipment than laboratory studies. This study compared the performance of two groups of listeners: ones participating in a sound booth with high-quality headphones and ones participating over the internet with whatever equipment was available. Listeners participated in a series of tasks rating children's productions of the sounds /s/, /S/, /Į/, /d/, /g/, /t/, and /k/. Using a variety of dependent measures, no group differences were found between internet and laboratory listeners. The potential utility of these judgments for studies of phonological development are discussed.

Full Paper

Bibliographic reference.  Munson, Benjamin (2013): "Assessing the utility of judgments of children's speech production made by untrained listeners in uncontrolled listening environments", In INTERSPEECH-2013, 2147-2151.