Logopedics, Vol. 5 No. 1, 2015.
Professional paper
Updates in Croatian Corpus of Child Language
Antonia Ordulj
orcid.org/0000-0002-1162-6223
; Sveučilište u Zagrebu, Filozofski fakultet, Odsjek za kroatistiku
Gordana Hržica
; Sveučilište u Zagrebu, Edukacijsko-rehabilitacijski fakultet, Odsjek za logopediju
Abstract
Croatian corpus of child language (CCCL – Kovačević, 2002) consists of language samples of spontaneous interaction of three children and adult speakers in their family environment. The corpus was created by longitudinal monitoring during the period of children's language acquisition from the onset of speech to approximately three years. CCCL is part of the Child Language Data Exchange System CHILDES (http://childes.psy.cmu.edu/data/Slavic/Croatian) and is encrypted in accordance with the CHAT system. In recent years the existing CCCL was subjected to a detailed analysis in order to make it consistent with the changes and developments of the CHILDES system. The aim of this paper is to give an overview of the encoding of CCCL in accordance with the rules of the CHAT system and to gain an insight into the changes that are primarily related to systematic coding of errors. During this process, transcripts have been connected with audio records by using the transcriber method described within the CHILDES system. All this allows for the easier access to the CCCL in general (which can now be viewed without installing specialized applications), and access to sound files of each transcript, available both as whole or related to a specific utterance.
Keywords
CHILDES; HKDJ; CHAT; coding; connecting audio records
Hrčak ID:
140212
URI
Publication date:
28.6.2015.
Visits: 2.569 *