KYNMAW — Khasi Bhashini Multi-Task Dataset

KYNMAW — Khasi Bhashini Multi-Task Dataset The largest open multi-task speech + text dataset for Khasi, an Austroasiatic language spoken by roughly 1.5 million people in Meghalaya, Northeast India — built to help bring Khasi into modern speech recognition, text-to-speech, and machine translation systems as part of If you’re working on low-resource ASR, endangered-language NLP, Indic speech translation, or Northeast Indian languages, this dataset is for you.… See the full description on the dataset page:

Source: Hugging Face Hub (RonitMehta260704/kh-en-dataset). Metadata imported from the dataset’s Hub tags.