Text to Speech Model
Gemini 3.8 Flash TTS
Text to Speech by Google
Publisher
Google
TypeText to Speech
Context Window6,000 tokens
ReleasedSeptember 2026
Input$0.50/MTok
Output$9.00/MTok
Ready to build with Gemini 3.8 Flash TTS?
Get Started FreeConfiguration
Parameters & options
Max Response Size16,384 tokens
ModeSelect
VoiceSelectsingle only
Prebuilt voice to speak in.
Speaker 1 NameTextdialogue only
The name speaker 1's lines start with in the script ("Name: line").
Speaker 1 VoiceSelectdialogue only
Speaker 2 NameTextdialogue only
The name speaker 2's lines start with in the script ("Name: line").
Speaker 2 VoiceSelectdialogue only
Style InstructionPrompt
Optional direction for the delivery: tone, pace, accent or character (e.g. "warm and unhurried"). It shapes the voice and is never spoken. Inline tags in the text itself, like <laugh>, <sigh> or <whispers>, act at that point.
Related models
Explore similar models
Start building with Gemini 3.8 Flash TTS
No API keys required. Create AI-powered workflows with Gemini 3.8 Flash TTS in minutes — free.