Fact · Languages and accents
12.5hours of Hindi audio
The LAHAJA Hindi benchmark contains 12.5 hours of Hindi audio from 132 speakers across 83 districts of India.
Checked against its source on October 17, 2026
In the source's own words
“with a total of 12.5 hours of Hindi audio, sourced from 132 speakers spanning 83 districts of India”
LAHAJA: A Robust Multi-accent Benchmark for Evaluating Hindi ASR Systems, Tahir Javed et al., arXiv preprint (AI4Bharat, IIT Madras), 2024, Abstract.
Open the original- What it measures
- LAHAJA is a test set, not a product. It mixes read speech with extempore (unscripted) speech on many topics, so it measures Hindi as it is spoken across regions.
- What it does not show
- Twelve and a half hours is a small test set. It shows how accents affect recognition, not an overall accuracy score for any app.