Fact · Speech recognition
2500hours
By the time of its 2019 paper, Mozilla's Common Voice corpus held 2,500 hours of audio from over 50,000 contributors, across 38 languages collecting data.
Checked against its source on October 16, 2026
In the source's own words
“Over 50,000 individuals have participated so far, resulting in 2,500 hours of collected audio.”
Common Voice: A Massively-Multilingual Speech Corpus, Rosana Ardila et al., arXiv (LREC 2020), 2020, Abstract.
Open the original- What it measures
- Contributors read sentences aloud and others validate them, which is how the corpus scales.
- What it does not show
- A November 2019 snapshot; the corpus has grown since.