Audio / Text-to-Speech

Ming-3.0-tts

Text-to-speech model from Ant Group · released Oct 2026

1182Speech Arena Elo (±13)
#23-29rank range of 105
2,640blind votes
3 voicesvoices in the arena
Oct 2026release

95% confidence interval: Elo 1169 – 1195. Models whose intervals overlap share a rank range.

How this is measured

Speech Arena (Text-to-Speech)

Elo rating (higher is better)

Models are ranked with an Elo system derived from blind user votes: listeners hear two speech samples generated from the same text and pick the one that sounds more natural. The provider-voice leaderboard uses each provider's own native voices. Each result is reported with a 95% interval and the number of votes (samples).

Release & sources

Release
Oct 2026 (month precision)
Release source
Artificial Analysis Speech Arena (month precision)