gg2
Gemini 3.8 text-to-speech released
letsdatascience.com

Gemini 3.8 text-to-speech released

The gist

Google released Gemini 3.8 Flash TTS and Flash-Lite TTS on September 23, 2026.
The models scale Gemini voices from 30 presets to over 2,000.

How it unfolded

  1. Sep 23, 2026Google introduced Gemini 3.8 Flash TTS and Gemini 3.8 Flash-Lite TTS, rolling out the same day in the Gemini API and Google AI Studio.
  2. Sep 23, 2026Launch pricing was listed at $0.50 per 1M text input tokens and $9.00 per 1M audio output tokens through December 31, 2026, rising to $1.00 and $18.00 from January 1, 2027, with a free tier also listed.
  3. Sep 23, 2026Google published benchmark tables showing Flash TTS at 71.4 overall on Hume AI's Text-to-Speech Voice Design Leaderboard (English), ahead of ElevenLabs Voice Design v3 at 70.8 and Inworld Voice Design at 69.8.

Sources

Read the full story in the app gg2 — free on the App Store