Audio / Text-to-Speech

GPT-Realtime-2

Text-to-speech model from OpenAI · released May 2026

1072Speech Arena Elo (±12)
#55-66rank range of 105
2,554blind votes
$191.56 / 1M charsAPI price
5 voicesvoices in the arena
May 2026release

95% confidence interval: Elo 1060 – 1084. 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
May 2026 (month precision)
Release source
Artificial Analysis Speech Arena (month precision)

More TTS models from OpenAI