Gemini 3.8 Flash TTS turns a script into a 24 kHz mono WAV in one of 30 voices, or two voices for a dialogue. Bracketed cues such as [whispering], [sighs] or [laughing] are acted out rather than read aloud.
Workbench
These examples were written for this endpoint. Replace the prompt and sample media URLs with your own accessible inputs, then check the current quote and model availability before submitting.
- 0:00 / –:––
whispered-confession Kore's voice with a single [whispers] cue at the start: the whole line drops into a close, breathy register without any style field. A bracketed cue is often all the direction a line needs.
- Kore
- 0:00 / –:––
storm-cabin-chapter The same romance narration as on the other pages, opened with a [softly] cue. Compare it with Gemini 2.5 Pro, which gets its pacing from an instructions field instead.
- Charon
- 0:00 / –:––
teasing-rival Puck's voice with [laughing] and [teasing] cues around a competitive line. Cues change the delivery mid-sentence; now and then the model reads a tag aloud, so listen before you ship.
- Puck
This example starts at $0.00032 per output second. The Playground updates the estimate when you change the settings.