Lahaja
AI4BharatLahaja is a Hindi speech recognition benchmark of 12.5 hours of read and extempore speech from 132 speakers across 83 districts of India, built to test models on regional accents. Each clip carries verbatim and normalized transcripts with the speaker's native language, district and occupation. AI4Bharat built it and released four NeMo baseline models with it.
Openness
3 medium confidence- license
- cc-by-4.0 (card License section)+mit (card metadata tag; conflicts with the License section)
- access
- auto (Hugging Face gate asks for contact information and approves automatically)
- dataset_card
- present (card and GitHub README describe the recordings and metadata fields)
Both licenses the card names allow commercial reuse with attribution, and the Hugging Face form approves requests automatically. The card should say which one governs.
- https://huggingface.co/api/datasets/ai4bharat/Lahaja recorded 2026-09-24
gated: auto; cardData license: mit
- https://huggingface.co/datasets/ai4bharat/Lahaja recorded 2026-09-24
License: mit (header); "This dataset is released under the CC BY 4.0."; "You need to agree to share your contact information to access this dataset"
- https://raw.githubusercontent.com/AI4Bharat/Lahaja/master/README.md recorded 2026-09-24
Metadata CSV fields: verbatim, normalized, native_language, native_district, occupation; links to models M1-M4
Adoption
1 high confidenceHugging Face downloads of the benchmark repository through the automatic gate. An evaluation set is downloaded once per evaluation setup, so downloads say little about how often scores on it are reported.
- https://huggingface.co/api/datasets/ai4bharat/Lahaja recorded 2026-09-24
232 downloads in the trailing 30 days for ai4bharat/Lahaja
Capability
not assessedLahaja is a speech test, and the map has no agreed way yet to turn hours of recorded speech into a measure of test depth, so it shows no capability score for now. Its overall score reflects adoption alone.
- https://arxiv.org/abs/2408.11440 recorded 2026-11-08
"a total of 12.5 hours of Hindi audio, sourced from 132 speakers spanning 83 districts of India"
- https://arxiv.org/pdf/2408.11440 recorded 2026-11-08
"a total of 12.5 hours of Hindi audio, sourced from 132 speakers"; read speech, digital interactions with voice assistants and "Extempore conversations" recorded by the participants; "All transcripts are generated in the native Devanagari script of Hindi"; "the initial transcript (maker) is verified by a senior transcriber (checker)"
- https://huggingface.co/datasets/ai4bharat/Lahaja recorded 2026-11-08
usage example loads the benchmark with split= "test"; "a total of 12.5 hours of Hindi audio, sourced from 132 speakers spanning 83 districts of India"
Verified 2026-09-24