Gemini 3.8 TTS Playground
Willison built a bring-your-own-key playground for Gemini 3.8’s new text-to-speech models.
Google’s new Gemini 3.8 TTS models include more than 2,000 voices and support custom voices from a 30-second sample. Willison says the API makes multi-speaker conversations straightforward, with separate voices and style instructions per character. His demo generated 1 minute 18 seconds of audio in about 20 seconds using Gemini 3.8 Flash TTS, at a cost of 2.74 cents. Simon Willison's note
Google’s new Gemini 3.8 TTS models include more than 2,000 voices and support custom voices from a 30-second sample. Willison says the API makes multi-speaker conversations straightforward, with separate voices and style instructions per character. His demo generated 1 minute 18 seconds of audio in about 20 seconds using Gemini 3.8 Flash TTS, at a cost of 2.74 cents. Simon Willison's note
score 7