Pnar-English Parallel Dataset

About Dataset

The Pnar English Parallel Dataset is a curated bilingual corpus containing 6,034 high quality sentence pairs translated by human contributors. Pnar also known as Jaintia is an indigenous language primarily spoken in the Jaintia Hills region of Meghalaya India. This gold standard dataset provides aligned English and Pnar text for machine translation multilingual language models cross lingual retrieval and other natural language processing applications. Each sentence pair has been manually reviewed and duplicate entries removed to ensure high data quality. Beyond AI research the dataset contributes to the documentation preservation and digital inclusion of the Pnar language and supports the development of language technologies for underserved communities. We thank CivicDataLab for their guidance and capacity building support on AI ready language datasets which informed the preparation and documentation of this resource.