GitHub - google-research-datasets/dakshina: The Dakshina dataset is a collection of text in both Latin and native scripts for 12 South Asian languages. For each language, the dataset includes a large collection of native script Wikipedia text, a romanizat

We publicly released a dataset last month and just now getting around to announcing it here. The Dakshina dataset is a collection of text in both Latin and native scripts for 12 South Asian languages:
1/5

github.com ↗