Dear All
Apologies for cross-postings.
I am involved in a project looking at intonation in Williams syndrome. We
are using a computer programme called PEPS-C to collect some of the data,
but also have a free speech task involving children producing a story based
on a picture book, _Frog Where Are You?_ - this book is commonly used in
studies of atypical speech.
We have decided to focus on nucleus placement, tone type and chunking, but
have run into a bit of a problem as we can't decide how to choose which
bits of the data to use so as to have some kind of comparability across the
speakers.
Some of the children have produced a story lasting around 90 seconds,
whereas others have done almost 5 minutes. I was originally thinking about
using 2 mins of speech for each speaker, as 90 secs isn't really enough,
but this would be problematic where we only have 90 seconds. We also
thought about focusing on the same e.g. 6 or 8 pages, but again we run into
problems where children haven't produced much on those pages.
Does anyone have any suggestions? I feel a bit silly about this as there
must be a sensible answer staring me in the face, but at present we haven't
been able to decide.
Awaiting your responses!
Jane
--
Dr Jane Setter
Director, English Pronunciation Research Unit
Department of Applied Linguistics
University of Reading
Whiteknights, PO Box 218
Reading, RG6 6AA
Tel: +44 (0)118 378 6089
Fax: +44 (0)118 975 3365
http://www.rdg.ac.uk/app_ling/setter.htm
|