Gemini 3.1 Flash TTS – with directed prompts

9 hours ago

Google released Gemini 3.1 Flash TTS, a new text-to-speech model controllable via prompts and accessible via the Gemini API.
The model requires detailed audio profile prompts, as demonstrated by the 'Jaz R.' example specifying scene, style, pace, and accent.
Users can customize output by modifying prompt details, such as changing accents from London to Newcastle or Exeter.
Gemini 3.1 Pro was used to code a UI for experimenting with the TTS model, highlighting its interactive potential.

Hasty Briefsbeta