Exploring the Impact of Centralised Translation Memory in OmegaT Translation Software The Case of the Xitsonga Language

Abstract

The role of Information Communication Technology (ICT) is crucial in language development. As the technological era evolves, it is important for the language space to keep up with current trends. Computer Assisted Translation (CAT) tools are still inadequately designed for African indigenous languages, which hinders their precision, usage, and digital participation, ultimately perpetuating linguistic disparities. Using a corpus-based translation approach, the study aims to explore the impact of centralised translation memory in OmegaT translation software, specifically in the Xitsonga language. CAT applications are used to enhance, accelerate, enrich, and deepen language skills. A qualitative research approach was adopted which comprised desktop research of secondary materials. A comparative analysis was used for analysing data as it involves a direct comparison that methodically examines the OmegaT translation memories and translated documents to identify their similarities and differences. Inconsistencies were noted in the terminologies used by different translators due to the absence of centralisation of translation memories and glossaries. The lack of translation equivalents in the standardised terminology list indicates that the terms have not been included for development; it is essential to welcome new words into the corpus and offer definitions for these terms to support the evolution of language and terminology.  Centralised translation memory could lead to reduced translation time and quicker turnaround for translation projects. It is recommended that OmegaT should include a sharing feature or a repository for storing translated terms, and this should be facilitated by Department of Sport, Arts and Culture, as they carry the mandate.