kapturecx/Chaashini
Chaashini (चाशनी) Chaashini — Hindi/Urdu for sugar syrup — is a continuously growing corpus of clean, single-speaker, studio-grade Indian-language speech built for training speech models (text-to-speech, speech recognition, speech language models). Every clip in the corpus has passed a strict multi-stage quality gate; the aim is purity over volume. Total: 1,390,417 clips · 2952.66 hours · 33 languages Format: mono 24 kHz FLAC (audio column) with a verbatim transcript and rich… See the full description on the dataset page: https://huggingface.co/datasets/kapturecx/Chaashini.
This repository is gated. The listing is public, but downloading a file means accepting the publisher’s terms at Hugging Face first — the links above take you there rather than around it.