Updated daily at 05:00 UTC
Benchmark
Artificial Analysis Text to Speech Arena
Measures: Listeners hear two blind text-to-speech clips of the same text and pick the more natural one; models are ranked by Elo computed from these votes, with 95% confidence intervals.
- Maintainer
- Artificial Analysis
- Introduced
- —
- Status
- Active
- Saturation
- Headroom
- Typical provenance
- Independent
- In scoring set
- Yes
Leaderboard
Independent and third-party results only.
No independent or third-party results yet. Show all results