speed-tb/bodo

Speed-Tb Phase 1 Bodo Narration Dataset Description The Bodo Speech Dataset, developed as part of the Speech Datasets and Models for Tibeto-Burman Languages (Project SpeeD-TB), funded under Mission Bhashini, is a transcribed speech corpus of the language. The full dataset comprises over 200 hours of high-quality audio recordings paired with accurate transcriptions in both IPA and Roman script, making it ** one of the largest speech resources for the language**… See the full description on the dataset page: https://huggingface.co/datasets/speed-tb/bodo.