G
Gemini 3.8 Flash-Lite TTS
Gemini 3.8 Flash-Lite TTS, a text-to-speech model from Google.
speechModelOverride: { model: "gemini-3.8-flash-lite-tts" }
Google·Released September 2026Pricing
- Input
- $0.50 / 1M
- Output
- $6.00 / 1M
Specifications
- Context window
- 6,000 tokens
- Max output
- 16,384 tokens
- Released
- September 2026
- Capabilities
- Low costMultilingual
Parameters
Model options, passed in
speechModelOverride.config.| Parameter | Type | Default | Range / options |
|---|---|---|---|
$mode | Select | single | singledialogue |
voicePrebuilt voice to speak in. | Select | Kore | ZephyrPuckCharonKoreFenrirLeda+24 |
speaker1_nameThe name speaker 1's lines start with in the script ("Name: line"). | Text | — | — |
speaker1_voice | Select | — | ZephyrPuckCharonKoreFenrirLeda+24 |
speaker2_nameThe name speaker 2's lines start with in the script ("Name: line"). | Text | — | — |
speaker2_voice | Select | — | ZephyrPuckCharonKoreFenrirLeda+24 |
styleInstructionOptional 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. | Prompt | — | — |
Voices · 30
Use this model
await mindstudio.textToSpeech({ text: "...", speechModelOverride: { model: "gemini-3.8-flash-lite-tts", config: { voice: "Zephyr", speaker1_voice: "Zephyr", }, }, });