Development and Evaluation of an English-to Igala Neural Machine Translation System using Deep Learning

Abstract

Low-resource languages face significant challenges in the digital age due to limited computational tools and data resources. This study presents the development of a neural machine translation (NMT) system for English-to-Igala translation using a Recurrent Neural Network (RNN) model. Igala is one of the under-resourced languages spoken in Nigeria. A bilingual parallel corpus of 1000 English-Igala sentence pairs was compiled and preprocessed to train and evaluate the system. The model achieved high translation accuracy as evidenced by BLEU scores above 0.5 on most test sentences. This research provides a foundational step for the development of computational resources for Igala and supports the broader goal of linguistic inclusivity in artificial intelligence.