The translation of the government platform VLibras and the accessibility of deaf people in higher education

In this article, based on Quadros & karnopp (2004) and Brito (1995), we analyzed the semi-automatic translations of the VLibras platform. We considered the syntax adequation of the translation from Portuguese to Libras through scientific articles from different areas, focusing on the challenges of access and permanence of death students on higher education. A qualitative methodology was used to analyze the corpus and describe the adequation of the grammatical elements of the oral hearing language to the spatial visual language. In order to analyze the automatic translation process, we proposed the following division of linguistic analysis: (i) verbal inflection; (ii) synthetic organization; (iii) punctuation. As a result, we found that there is a pattern on the translation that generates mistakes. We understand that most errors from the automatic translation were due to the lack of consideration of the grammatical aspects of Libras concerning Portuguese. Changing this scenario requires a deeper understanding of studies such as this, aiming to investigate the errors, the hits, and search solutions to improve automatic translation tools.