Humanness Index™ · TTS model
by Cartesia
Sonic 2 was Cartesia's second generation voice model, announced in early 2025 alongside the company's $64M Series A.
Standings as of Jun 13, 2026, 01:49 UTC
A real arena clip: a cloned source voice reading a customer support prompt at phone quality.
Sonic 2 was Cartesia's second generation voice model, announced in early 2025 alongside the company's $64M Series A. It cut time to first audio to roughly 90 ms, added a Turbo variant that reached 40 ms, and supported 15 languages. It established Sonic as one of the fastest production text to speech APIs before Sonic 3 took over as the flagship.
Sources: cartesia.ai
The fastest production TTS of its moment, with a 40 ms Turbo variant. In our 50 trial streaming benchmark Sonic 2 returned first audio in a median of 159 ms including network time.
Sources: docs.cartesia.ai
Standings as of Jun 13, 2026, 01:49 UTC
| Rank | Provider | Model | Humanness | Latency |
|---|---|---|---|---|
| #16 | ElevenLabs | Multilingual v2 | 59 | 1006 ms |
| #17 | ElevenLabs | Flash v2 | 59 | 226 ms |
| #18 | Cartesia | Sonic 2 | 44 | 159 ms |
| #19 | Cartesia | Sonic 3 | 27 | 166 ms |
| #20 | Gradium | Gradium TTS | 24 | 332 ms |
The benchmark is open source. Suggest a model, read the methodology, or ask us to put your voice in the arena.
