Vedavani is the first benchmark dataset for automatic speech recognition (ASR) on Vedic Sanskrit poetry, consisting of richly annotated verses from the Rig Veda and Atharva Veda. This corpus captures the unique prosodic structure, phonetic complexity, and chanting style found in traditional Vedic recitation.
🔗 Paper: Vedavani: A Benchmark Corpus for ASR on Vedic Sanskrit Poetry (ACL 2025)📁 GitHub Repository:… See the full description on the dataset page:
https://huggingface.co/datasets/sanganaka/Vedavani-Dataset.
