armenian_gold_dataset
Narek889/armenian_gold_dataset
~181k English–Armenian parallel sentence pairs for machine translation (en↔hy). Compiled from OPUS, Tatoeba, Wikipedia, Western Armenian corpus and synthetic data.
machine-translationparallel-corpusarmenian-nlpenglish-armenianlow-resource-languagedatasethuggingfacesynthetic-data
Summary
This repository provides a ~181k English–Armenian parallel sentence dataset for machine translation, hosted on Hugging Face. It combines OPUS, Tatoeba, Wikipedia, Western Armenian corpus, and synthetic technical data, with documentation for NMT fine-tuning. It is useful but appears to be a compiled dataset rather than a maintained codebase, with some license ambiguity and incomplete source accounting.