Building a Functional Machine Translation Corpus for Kpelle

TL;DR

This paper introduces the first publicly available English-Kpelle machine translation dataset, demonstrating effective fine-tuning of a multilingual model and highlighting its potential for advancing NLP in low-resource African languages.

Contribution

It provides the first English-Kpelle dataset for machine translation and shows how fine-tuning improves translation quality, enabling broader NLP applications for low-resource languages.

Findings

Achieved BLEU scores of up to 30 in Kpelle-to-English translation.

Demonstrated the dataset's utility for various NLP tasks.

Aligned results with benchmarks for other African languages.

Abstract

In this paper, we introduce the first publicly available English-Kpelle dataset for machine translation, comprising over 2000 sentence pairs drawn from everyday communication, religious texts, and educational materials. By fine-tuning Meta's No Language Left Behind(NLLB) model on two versions of the dataset, we achieved BLEU scores of up to 30 in the Kpelle-to-English direction, demonstrating the benefits of data augmentation. Our findings align with NLLB-200 benchmarks on other African languages, underscoring Kpelle's potential for competitive performance despite its low-resource status. Beyond machine translation, this dataset enables broader NLP tasks, including speech recognition and language modelling. We conclude with a roadmap for future dataset expansion, emphasizing orthographic consistency, community-driven validation, and interdisciplinary collaboration to advance inclusive language technology development for Kpelle and other low-resourced Mande languages.