Sicilian Translator: A Recipe for Low-Resource NMT
arXiv:2110.01938
Abstract
With 17,000 pairs of Sicilian-English translated sentences, Arba Sicula developed the first neural machine translator for the Sicilian language. Using small subword vocabularies, we trained small Transformer models with high dropout parameters and achieved BLEU scores in the upper 20s. Then we supplemented our dataset with backtranslation and multilingual translation and pushed our scores into the mid 30s. We also attribute our success to incorporating theoretical information in our dataset. Prior to training, we biased the subword vocabulary towards the desinences one finds in a textbook. And we included textbook exercises in our dataset.
arXiv comment7 pages, 2 tables
References in corpus (6)
- Beyond English-Centric Multilingual Machine Translation
- Massively Multilingual Neural Machine Translation in the Wild: Findings and Challenges
- Sockeye: A Toolkit for Neural Machine Translation
- Google's Multilingual Neural Machine Translation System: Enabling Zero-Shot Translation
- Revisiting Low-Resource Neural Machine Translation: A Case Study
- Investigating Multilingual NMT Representations at Scale