speech to text benchmark
Common Voice 15
Common Voice is a massively-multilingual collection of transcribed speech intended for speech technology research and development. Version 15.0 contains 28,750 recorded hours across 114 languages, consisting of crowdsourced voice recordings with corresponding transcriptions.
Updated Aug 7, 2026
Published models1
Registry coverage1
MetricScore
EvidenceB
Common Voice 15 leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | AC | 0.076 points | 100.0% | 1 | C |
Common Voice 15 highlights
The top published results on this benchmark's own scale.
What is Common Voice 15?
Definition and scoring fields from the benchmark registry.
Common Voice is a massively-multilingual collection of transcribed speech intended for speech technology research and development. Version 15.0 contains 28,750 recorded hours across 114 languages, consisting of crowdsourced voice recordings with corresponding transcriptions.
Scores are shown in points. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- Common Voice 15
- Modality
- audio
- Primary category
- speech to text
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- common-voice-15|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about Common Voice 15.
Which model scores highest on Common Voice 15?
Qwen2.5-Omni-7B is currently ranked first with 0.076 points.
What does Common Voice 15 measure?
Common Voice is a massively-multilingual collection of transcribed speech intended for speech technology research and development. Version 15.0 contains 28,750 recorded hours across 114 languages, consisting of crowdsourced voice recordings with corresponding transcriptions.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
1 unique published model results are currently shown.
Does this benchmark affect the overall score?
This benchmark is preserved as source-native evidence but is not eligible for the current overall score.