Preprint
IndicVoices: Towards building an Inclusive Multilingual Speech Dataset for Indian Languages
Tahir Javed et al., arXiv preprint (AI4Bharat, IIT Madras), 2024
Read the originalRead on October 17, 2026
- What it is
- Describes IndicVoices, a speech dataset of 22 Indian languages collected by AI4Bharat. Most of the audio is extempore or conversational speech rather than read text. The abstract gives the total hours, speakers and speech styles.
- How much weight it carries
- A preprint from the group that built the dataset. Later catalogue pages give different totals, so this guide uses only the figures in the 2024 abstract.
Facts taken from this source
- The IndicVoices dataset contains 7,348 hours of read, extempore and conversational speech from 16,237 speakers, covering 22 Indian languages.
“a total of 7348 hours of read (9%), extempore (74%) and conversational (17%) audio from 16237 speakers covering 145 Indian districts and 22 languages” (Abstract)
- The IndicVoices authors built IndicASR, which they describe as the first speech recognition model to support all 22 languages in the Eighth Schedule of the Indian Constitution.
“we build IndicASR, the first ASR model to support all the 22 languages listed in the 8th schedule of the Constitution of India” (Abstract)