Sicilian Translator: A Recipe for Low-Resource NMT

5 Oct 2021  ·  Eryk Wdowiak ·

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.

PDF Abstract

Datasets


  Add Datasets introduced or used in this paper

Results from the Paper


Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Machine Translation Arba Sicula Larger BLEU (En-Scn) 35.0 # 1
BLEU (Scn-En) 36.8 # 1
Machine Translation Arba Sicula Many-to-Many BLEU (It-Scn) 36.5 # 1
BLEU (Scn-It) 30.9 # 1

Methods