Gemini 3.1 Flash TTS for Text to Speech:

ABOUT THE WORKFLOW

Type a Script, Get Speech
Type the text you want spoken, pick one of 30 voices, and the model reads it back as audio. Describe the tone, accent, and pacing in plain words to direct the performance. Name two speakers for a back and forth conversation, all in one clip.

Partner node. This workflow calls an external API, so each run uses credits from your API wallet. No API key needed. Floyo handles the connection.

Model

• Gemini 3.1 Flash TTS by Google. A text-to-speech model strong at expressive, natural delivery, with control over style, accent, and pacing across 70 plus languages.

HOW IT WORKS

Step 1. Write your text
The words you want spoken. Anything from a single line to a full script.
Works great with: narration · ads · explainers · audiobooks

Step 2. Pick a voice
Choose from 30 voices, each with its own character. The default reads cleanly for most scripts.

Step 3. Describe the delivery (optional)
Direct the tone, accent, and pacing in plain words, like "warm documentary narrator, slow and deliberate" or "upbeat radio host."

Step 4. Set two speakers (optional)
Name two speakers and give each a voice for a back and forth conversation. Leave it empty for a single narrator.

Step 5. Hit run and download
The model returns spoken audio. Preview it in the workflow, then download the file.
Ready for: Premiere · Audacity · CapCut · any editor

First time? Leave every setting as-is. The defaults are the right starting point for almost everyone.

RECOMMENDED SETTINGS

Quick-start guide. Find the goal that matches yours and copy the settings.

• Standard voiceover (most people) — 1 voice · Auto language · temperature 1 · mp3. The right starting point for almost everyone.

• Want a two-person dialogue — Name speaker 1 and speaker 2 and assign each a voice. The model voices both in one clip with consistent characters.

• Want a specific tone or accent — Write a style instruction describing who is speaking and how, such as "calm British narrator, measured pace."

• Want steadier, more repeatable delivery — Lower the temperature. A lower value keeps the read consistent between runs, higher adds variation.

• Need a different file type — Switch the output format to match where the audio is going.

• The delivery is not landing — Make the text itself emotionally clear and keep the style instruction pointed in the same direction as the script.

Prompt: Write the style instruction like a director's note: who is speaking, the scene, and the mood. "A 1940s hardboiled detective narrating a late-night radio show, dry and deadpan" gives the model far more to work with than "read this dramatically." You can also drop inline tags into your text, like [whispers] or [long pause], to shape specific moments.

LEARN

📹 Videos

• Intro to Floyo

• ComfyUI 101 Free Course ft. Sebastian Kamph

• Floyo 101 for Team Collaboration

✨ Quick links

• Help docs

• LoRA Training guide

• Discover more workflows

USE CASES

🎬 Video Creators
Generate voiceover for explainers, ads, and social clips without booking a voice actor or recording session.

🎧 Podcasters
Voice a two-person conversation from a script, with a distinct, consistent voice for each speaker in a single clip.

📚 Audiobook & Long-Form
Read chapters and articles in a directed tone, using pacing notes and pauses to match the material.

🌍 Localization Teams
Produce spoken audio across 70 plus languages and regional accents from one model, instead of juggling a vendor per region.

WHAT WORKS BEST / WHAT TO AVOID

✅ Works great

• Clear scripts with natural punctuation

• A specific style note: who is speaking and how

• Emotionally evocative text that matches the tone you want

• Inline tags for pauses, emphasis, and emotion

⚠️ May produce softer results

• A vague style note like "make it sound good"

• A style instruction that fights the mood of the text

• Overspecifying every detail instead of leaving room

• Very long single blocks with no pacing cues

FAQ

What is Gemini 3.1 Flash TTS?
Gemini 3.1 Flash TTS is Google's text-to-speech model in the Gemini 3.1 Flash family. It converts written text into natural, expressive speech with control over voice, style, accent, and pacing. It supports 30 prebuilt voices and 70 plus languages, and it can voice a two-person dialogue in a single generation.

How do I control the tone and accent of the voice?
You write a style instruction in plain language describing who is speaking and how, such as "energetic sports commentator" or "soft, slow bedtime story reader." You can also embed inline tags in your text, like [whispers], [excited], or [long pause], to direct specific words and moments. The style note and the text content work best when they point in the same direction.

Can it generate a conversation between two people?
Yes. Name two speakers, assign each a voice, and label your lines by speaker. The model produces one audio clip with distinct, consistent voices for each character, so you do not have to generate separate takes and stitch them together.

How many voices and languages does it support?
There are 30 prebuilt voices to choose from and support for more than 70 languages, including regional accent variants. Language detection runs automatically from your text, or you can set the language explicitly for the most reliable result.

Do the outputs have a watermark?
Yes. Google embeds an inaudible SynthID watermark in the audio to mark it as AI-generated. It does not change how the audio sounds, but it stays in the file.

Can I use the results commercially?
Gemini 3.1 Flash TTS is a proprietary Google model, so commercial use is governed by Google's terms rather than an open license. Commercial use is generally allowed, but review Google's current terms and note that outputs carry the SynthID watermark.

How to run Gemini 3.1 Flash TTS online?
You can run Gemini 3.1 Flash TTS online through Floyo. No installation, no setup, no API key to wire up. Open the workflow in your browser, type your text, pick a voice, and hit run. Free to try.

WHY FLOYO?

Floyo is the only platform with team collaboration for ComfyUI in the browser. You run workflows with no install. You share run history, assets, and models across your team. You pay only when you generate. Floyo supports open-source and closed-source models.

A creator runs a voiceover and likes the take. A teammate opens that exact run from shared history and keeps going. No file handoffs. No version confusion.

For studios and enterprise teams, Floyo adds private workspaces, pooled resources, and a team usage dashboard. Other ComfyUI cloud tools run for one person at a time. Floyo runs for the whole team, with transparent per-generation costs.

Ready to try it?
Type a script, pick a voice, and run it. The settings are already set.

→ Launch Workflow, Free

Questions? Watch the free course or check the FAQ above.