Data Augmentation to Low-Resource Languages Approach: Galician case

Abstract

In this Final Thesis, two proposals for data augmentation have been made for the training of two bilingual translation models in Spanish-Galician and English-Galician from parallel corpora of Portuguese. One of the proposals for adapting Portuguese to Galician has been transliteration and the other translation with Apertium RBMT. The results obtained have been promising in the English-Galician pair and leave the door open for future research.