# Gemini TTS Gemini TTS is Google's text-to-speech family. Gemini 3.8 Flash TTS is Google's flagship creative voice model and Gemini 3.8 Flash Lite TTS its fast, cost-efficient tier, both generally available since 22 September 2026. In Sloosh both run in Spacelab's Text to Speech node, with 30 voices, a separate Style note and scripts up to 8,192 characters. Web page: https://sloosh.ai/models/gemini-tts [How to use Gemini TTS](https://sloosh.ai/docs/models#audio-models) ## Key facts in Sloosh | | Gemini 3.8 Flash TTS | Gemini 3.8 Flash Lite TTS | | --- | --- | --- | | Made by | Google | Google | | Where | Spacelab, Text to Speech node | Spacelab, Text to Speech node | | Makes | speech audio | speech audio | | Inputs | script text; voice; Style (a separate delivery note) | same | | Voices | 30 prebuilt voices (default Kore) | the same 30 voices | | Languages (Google) | over 130, detected from the script | over 100, detected from the script | | Language picker | none | none | | Script limit | 8,192 characters | 8,192 characters | ## The models ### Gemini 3.8 Flash TTS Google calls it its "flagship creative text-to-speech model for studio-grade voice fidelity, expressive acting, and regional dialects across 130 languages" (Gemini API docs: Models). Use it when the read has to carry emotion. ### Gemini 3.8 Flash Lite TTS Google's "fast, cost-efficient text-to-speech model built for high-throughput production workloads" (Gemini API docs: Models), built to replace Gemini 3.1 Flash TTS Preview (Gemini API changelog). Use it for many lines or drafts. The exact price of each shows in Spacelab before you run. ## Best for - Narration where tone is set apart from the words, through the node's Style field (Gemini API docs: Speech generation). - Scripts in many languages, detected automatically (Gemini API docs: Speech generation). - Voiceovers for ads and explainers that you then put under a video or lip-sync to a face. ## How to use Gemini TTS in Sloosh 1. Open Spacelab ([sloosh.ai/spacelab](https://sloosh.ai/spacelab), signed in) and add a Text to Speech node (Audio). 2. In the Model picker, choose Gemini 3.8 Flash TTS or Gemini 3.8 Flash Lite TTS. 3. Wire a Text Input, Prompt or Any LLM node into the script (text) port. 4. Pick one of the 30 voices and, if needed, write a short delivery note in Style, for example "Calm and warm, like a late-night radio host". 5. Run the node. The exact price shows in Spacelab before you run. 6. Wire the audio into Merge audio + video, or into Lipsync to make a face say it. ## Gemini TTS prompting tips - Keep the script a word-for-word transcript; put sustained delivery notes (emotion, pace, tone) in Style (Gemini API docs: Speech generation). - Capitalize a word to stress it (Gemini API docs: Speech generation). - Use short inline tags such as or ; Google recommends English tags even in non-English scripts (Gemini API docs: Speech generation). - Avoid long style blocks and meta-instructions about keeping the voice steady; Google calls them the most common cause of voice drift (Gemini API docs: Speech generation). ## Gemini TTS example prompts **Prompt 1** ```text Style: bright and upbeat. Script: Big news! Our summer collection drops this Friday. ``` **Prompt 2** ```text Style: calm and warm, like a late-night radio host. Script: Close your eyes. Let the day go. Breathe in. ``` **Prompt 3** ```text Style: (empty). Script: Here is how to set up your new speaker in under a minute. ``` ## How it compares in Sloosh | Model | Script limit | Voices | Direction | | --- | --- | --- | --- | | Gemini 3.8 Flash TTS | 8,192 chars | 30 | Style field | | Gemini 3.8 Flash Lite TTS | 8,192 chars | 30 | Style field | | ElevenLabs v4 | 5,000 chars | 21 | inline audio tags | | ElevenLabs Multilingual v2 | 10,000 chars | 21 | language picker, 29 languages | | MiniMax Speech 2.8 HD | 10,000 chars | 314 | language picker, 36 languages | From Sloosh's catalog rows. ## Where else it runs - Gemini API (Gemini 3.8 Flash TTS, Gemini 3.8 Flash Lite TTS) (Gemini API docs: Models, Oct 2026). ## Frequently asked questions ### What is Gemini TTS? Gemini TTS is Google's text-to-speech in the Gemini API. The current models are Gemini 3.8 Flash TTS and Gemini 3.8 Flash Lite TTS, both GA on 22 September 2026 (Gemini API changelog). ### What is the difference between Flash TTS and Flash Lite TTS? Flash TTS is Google's flagship for expressive, studio-grade voices; Flash Lite TTS is the fast, cost-efficient tier (Gemini API docs: Models). In Sloosh both have the same 30 voices and 8,192-character limit. ### How many languages does Gemini TTS support? Over 130 for Flash TTS and over 100 for Flash Lite TTS, detected from the script (Gemini API docs: Speech generation). Sloosh's node has no language picker for these models. ### How do I control the tone of the voice? Write a short note in the node's Style field, and capitalize words you want stressed (Gemini API docs: Speech generation). ### How long can the script be in Sloosh? Up to 8,192 characters for both models. ### Can I make a face say the voiceover? Yes. Wire the Text to Speech output into a Lipsync node with a face video or photo. ## Related - Models: ElevenLabs v4, ElevenLabs Multilingual v2, MiniMax Speech 2.8 HD - Spaces: [Talking clip](https://sloosh.ai/spaces/talking-clip) - Docs: [Models](https://sloosh.ai/docs/models.md#audio-models) ## Sources - [Sloosh docs: Models (Audio models table)](https://sloosh.ai/docs/models.md) - [Sloosh Space: Talking Clip](https://sloosh.ai/spaces/talking-clip/agents.md) - [Gemini API docs: Speech generation](https://ai.google.dev/gemini-api/docs/speech-generation) - [Gemini API docs: Models](https://ai.google.dev/gemini-api/docs/models) - [Gemini API docs: Changelog](https://ai.google.dev/gemini-api/docs/changelog)