Corpus details

Child Language Data Exchange System

Official name:Child Language Data Exchange System
Common name:CHILDES
Language:Multilingual; English; Dutch
Language type:spoken
Corpus type:child language
Description:The CHILDES corpus is of a collection of data sets consisting of transcripts of young children's speech. In many cases, transcripts are available over a certain period of time, allowing research into first language acquisition. The data sets vary in size, annotation, setting, etc.
Exploration:CLAN is software that has been designed especially for the exploration of CHILDES material. However, the data are stored in ASCII (text) format and can therefore also be explored with Wingrep. WordSmith is less useful for the exploration of CHILDES material, because it ignores line feeds and does not show separate lines in the output..
Example:*JAC: en dan gaan we boekje kijken.
*SAR: daar paajt [: paard].
*JAC: een paard ["]?
*JAC: een schildpad.
*SAR: bed.
*JAC: een bed ["], ja.
*JAC: ze zijn lekker aan (he)t slapen, he.
*SAR: beer.
*JAC: een beer ["], ja.
*SAR: o [=! screaming] !
Location:Faculty network, folder G:\LET\Data\Corpora\Nederlands\Childes
Details:CHILDES is part of TalkBank, a multilingual corpus containing sample databases from within several subfields of communication, including first language acquisition, second language acquisition, conversation analysis, classroom discourse, and aphasic language.
See Also: 
Name:CHILDES website
Description:Home page of CHILDES

back to overview