Towards Santali Linguistic Inclusion: Building the First Santali-to-English Translation Model using mT5 Transformer and Data Augmentation

TL;DR

This paper demonstrates the feasibility of building a Santali-to-English translation model using mT5 transformer and data augmentation, addressing the low-resource challenge for this underrepresented language.

Contribution

It is the first to develop a Santali-English translation model leveraging transfer learning and data augmentation techniques in a low-resource setting.

Findings

mT5 transformer outperforms untrained models for Santali translation

Santali-English corpus yields better results than Santali-Bangla with mT5

Data augmentation improves translation performance

Abstract

Around seven million individuals in India, Bangladesh, Bhutan, and Nepal speak Santali, positioning it as nearly the third most commonly used Austroasiatic language. Despite its prominence among the Austroasiatic language family's Munda subfamily, Santali lacks global recognition. Currently, no translation models exist for the Santali language. Our paper aims to include Santali to the NPL spectrum. We aim to examine the feasibility of building Santali translation models based on available Santali corpora. The paper successfully addressed the low-resource problem and, with promising results, examined the possibility of creating a functional Santali machine translation model in a low-resource setup. Our study shows that Santali-English parallel corpus performs better when in transformers like mt5 as opposed to untrained transformers, proving that transfer learning can be a viable technique that works with Santali language. Besides the mT5 transformer, Santali-English performs better than Santali-Bangla parallel corpus as the mT5 has been trained in way more English data than Bangla data. Lastly, our study shows that with data augmentation, our model performs better.