Machine Translation in the Covid domain: an English-Irish case study for LoResMT 2021

TL;DR

This paper develops and evaluates domain-specific machine translation models for Covid-related English-Irish translation, demonstrating significant BLEU score improvements through dataset extension and domain adaptation techniques.

Contribution

It introduces a Covid-specific dataset and compares various fine-tuning approaches, showing the effectiveness of dataset extension for domain adaptation in translation models.

Findings

Extending the in-domain dataset by 5k lines improved BLEU score by 27 points.

The highest-performing model used a Transformer architecture trained on an extended Covid dataset.

Domain adaptation techniques enhanced translation quality for Covid-related content.

Abstract

Translation models for the specific domain of translating Covid data from English to Irish were developed for the LoResMT 2021 shared task. Domain adaptation techniques, using a Covid-adapted generic 55k corpus from the Directorate General of Translation, were applied. Fine-tuning, mixed fine-tuning and combined dataset approaches were compared with models trained on an extended in-domain dataset. As part of this study, an English-Irish dataset of Covid related data, from the Health and Education domains, was developed. The highest-performing model used a Transformer architecture trained with an extended in-domain Covid dataset. In the context of this study, we have demonstrated that extending an 8k in-domain baseline dataset by just 5k lines improved the BLEU score by 27 points.