Fact · Speech recognition
102languages
The FLEURS benchmark is a parallel speech dataset in 102 languages, with about 12 hours of speech supervision per language.
Checked against its source on October 17, 2026
In the source's own words
“FLEURS is an n-way parallel speech dataset in 102 languages built on top of the machine translation FLoRes-101 benchmark, with approximately 12 hours of speech supervision per language.”
FLEURS: Few-shot Learning Evaluation of Universal Representations of Speech, Alexis Conneau et al., arXiv (Computation and Language), 2022, Abstract.
Open the original- What it measures
- FLEURS is a multilingual benchmark. Each language has the same sentences, read aloud and recorded.
- What it does not show
- The sentences are read aloud, so the benchmark does not test spontaneous conversation or noisy rooms. Hours per language are approximate.
Source
Cited on
- Dictation Languages: What Each App and Device SupportsGuide
- Hindi Voice Typing and Speech to TextGuide
- Speech Recognition Benchmarks and LeaderboardsGuide
- Whisper Supported Languages and Accuracy by LanguageGuide
- Speech to Text in French, German, Russian and MoreGuide
- What are the best datasets for speech recognition?Question