Creating a Parallel Corpus for Machine Translation: A Case Study of Kru and Krio

In recent years, machine translation has seen significant advancements. Despite this progress, numerous low-resource languages still lack robust translation systems. The ubiquity of English in digital realms and daily interactions pose a substantial obstacle for Kru and Krio speakers, especially those who lack literacy skills. This challenge hampers their access to online content and effective communication. If not tackled proactively, these languages risk marginalization and exclusion from various platforms, potentially leading to their gradual decline in favour of English. To counteract this trend, the creation of comprehensive language resources is vital. Parallel corpora, among other resources, can play a crucial role in addressing this issue. This research proposes a proposed model aimed at constructing a parallel corpus for Kru and Krio, two languages spoken in Sierra Leone. The primary objective is to facilitate the development of machine translation systems tailored to these languages. The research paper introduces an approach for constructing the parallel corpus and delves into the potential challenges as well as opportunities inherent in this endeavour.