Brent corpus

[Brian MacWhinney]

Dear Info-CHILDES,
  I am happy to announce the addition to CHILDES of a large corpus of speech
from mothers to their preverbal infants.  This corpus was collected by
Michael Brent and Jeff Siskind and is probably now the largest collection of
transcripts fully linked to audio in CHILDES.  As I will note in my next
message, it is a particularly nice illustration of the new CHILDES feature
of a "Browsable Corpus".  Thanks to Michael and Jeff for this new corpus.
Here is the readme.  Further documentation on the details of there method
will be included in the manual and on the web.

--Brian MacWhinney



This corpus includes about 100 hours of recordings from mothers of preverbal
children. All of the transcripts are completely linked to the audio.
Recordings and transcripts are available for 16 mother-child pairs. 8 of
these were recruited without regard to the education level of the mother
but, as it turned out, all had at least a bachelor's degree. Subsequently, 8
more mother-child pairs were recruited with the requirement that the mother
must not have completed more than 2 years of college. The education levels
of each of the subjects can be viewed by clicking here. 6 of the 8
lower-education subjects completed all or almost all of the sessions, but
the remaining 2 provided only about 7 usable sessions each.



Each mother-infant pair was recorded in their home during approximately 14
sessions lasting 1.5 to 2 hours each. Sessions were separated by
approximately 2 weeks during the period when the child was between 9 and 15
months old. The mother and child were left alone during the recording
session. The middle 75 minutes of each session were extracted, digitized,
and transcribed. Some details of the methods can be found in our paper:


Brent, M. R. & Siskind, J. M (2001). The role of exposure to isolated words
in early vocabulary development. Cognition, 81, 31-44.


If you use these data, please cite that paper.