Fact · Speech recognition
The Common Voice paper says the corpus is the largest audio corpus in the public domain for speech recognition, by both hours and number of languages.
Checked against its source on October 17, 2026
In the source's own words
“To our knowledge this is the largest audio corpus in the public domain for speech recognition, both in terms of number of hours and number of languages”
Common Voice: A Massively-Multilingual Speech Corpus, Rosana Ardila et al., arXiv (LREC 2020), 2020, Abstract.
Open the original- What it measures
- The paper was written in late 2019, when the project had 2,500 hours of collected audio from more than 50,000 people in 38 languages. The corpus grows with each release, so the current totals are higher.
- What it does not show
- This is the authors' own claim at the time of writing. It describes the size of the corpus, not the quality of its transcripts, which are volunteer-validated.