Diseño de un corpus lingüístico.

Abstract

INTRODUCTION: The objective of the research was to design a linguistic corpus of children's language in children from 3 years 11 months to 4 years 11 months belonging to Child Development Centers in the municipality of Cúcuta. METHODS: A cross-sectional descriptive quantitative study was carried out, whose universe of study consisted of 92 children. Recordings were made for 8 weeks 3 days a week, in videos of 20 minutes in the classroom. The sample studied was 22 children, who underwent transcription through SALT software. RESULTS: Corpus of oral typology, consisting of 7,498 words, distributed in 22 texts, with a total of 4 hours and 20 minutes of recording and 22 participants (12 girls, 10 girls). ANALYSIS AND DISCUSSION: The main contribution of this corpus is the contribution of a linguistic resource that is still scarce today. This corpus facilitates a sample of the children's linguistic variety, both in audio and text formats. In addition, these texts are enriched with encodings, from which all types of information can be extracted automatically referring to the levels of syntax, semantics, discourse, speed, fluency, errors and omissions. CONCLUSIONS: The corpus complies with the characteristic of modern corpora of being an electronic format, allowing the storage and manipulation of data and its possible exchange with other interested researchers, but does not comply with the characteristic of representativeness, because diversity in terms of ranges old is low.