Transcription and data annotation using ELAN and LancsBox

Abstract

This study aims at demonstrating how the transcription tool ELAN 5.9 (2020) and the corpus analysis LancsBox 5.1.2 (BREZINA; WEILL-TESSIER; MCENERY, 2020) have been contributing to transcribe sociolinguistics interviews made at Grupo de Pesquisa em Linguagem, Interação e Sociedade (GELINS) as well as to the automatically extract variable linguistic phenomena. In order to do that, norms through which the interviews are transcribed, ways of using ELAN 5.9 (2020) for transcription, and the way to morphologically tag data and to search speech data for linguistic variable phenomena using LancsBox 5.1.2 (BREZINA; WEILL-TESSIER; MCENERY, 2020) are presented. The two tools have been useful to an aligned transcription, a morphological annotation, and an automated search on large amounts of speech data. This text contributes to the exploration of tools that enables a faster and more accurate transcription of speech data as well as more automated searches on large amounts of data.