Gemini 3.8 Flash TTS
by Google DeepMind
Google's voice for narration and dialogue.
Laughs, sighs and pauses land where you write them.

Key features
Technical specifications
Up to 2
Two distinct voices in one dialogue generation
10,000 chars
Maximum script length per speech generation
2
Gemini 3.8 Flash TTS and Gemini 3.8 Flash-Lite TTS
SynthID
Inaudible AI-provenance mark on every clip
Use cases

Video narration
Lay a narrator under an AI or live-action cut, then regenerate one line when the edit changes.

Character dialogue
Two-voice scenes for shorts, games and explainers, with each reply timed to the line before it.

Audiobooks and long-form
Work through a chapter in long passages with one cast voice, and place the breaths and pauses yourself.

Podcast intros and segments
A host and a guest in one generation, for cold opens, ad breaks and recap segments.

Localized voice-over
Run the same script in another language from the same voice picker, filtered to native voices.

High-volume reads
Use Flash-Lite for product reads, read-aloud and drafts, then render the keepers on Flash.
Prompt examples



Pause for effect
The door was already open. <short pause> Someone had been inside.


Product walkthrough
Tap the plus icon, pick a template, and your first page is ready.

Two-voice exchange
Ines: Did the parcel come? Theo: It did. <gasp> It's the wrong colour.
Overview
Gemini 3.8 Flash TTS is Google's text-to-speech model, launched on September 23, 2026 with a faster sibling, Gemini 3.8 Flash-Lite TTS. It reads a script in a voice from Google's voice library, performs the laughs, sighs and pauses you place inline, and voices scenes between two speakers.
What Gemini 3.8 Flash TTS does differently
Gemini 3.8 reads your script as a word-for-word transcript. Sounds go exactly where you put them, written in angle brackets such as <laugh> or <short pause>, and nothing outside them gets performed as an instruction. That makes a take easy to predict and easy to fix: move one tag and generate again. The trade is that tone comes from the voice you cast and the way the line is written, not from a note telling the voice how to feel.
Gemini 3.8 Flash TTS on Morphic
On Morphic, Gemini 3.8 Flash TTS runs in Speech for single-voice scripts of up to 10,000 characters and in Dialogue for two-speaker scenes. Pick it in the Audio model picker, choose voices from Google's library filtered by language and accent, and the audio lands on your Canvas. Gemini 3.8 Flash-Lite TTS uses the same voices at a faster, lower-cost setting. It replaced the retired Gemini 3.1 Flash TTS.
Simple pricing
Get started for free today, with the option to upgrade or cancel anytime.
Basic
2400 monthly credits
1 user only
All models
Workflows
Standard
3625 monthly credits
1 user only
All models
Workflows
Pro
6350 shared monthly credits
1 user
All models
Workflows
Max
24650 shared monthly credits
1 user
All models
Workflows
Enterprise
For higher limits
Custom
pricing and billing terms

Free
For playing around
$0
forever free
FAQs
Other models
Explore the rest of the Morphic model catalog.
Seedream 5.5
ByteDance
ByteDance's Seedream image model, released November 2026. Clean multi-round edits, lifelike realism, and sharper text.
Gemini Omni Flash 1.2
Google DeepMind
Google's video model, now fifteen seconds long. Edit the whole take for 30% less a second.
ChatGPT Images 2.6
OpenAI
OpenAI's image model, released November 2026. Animated GIFs, SVG export, faster drafts.
SAM 3.1
Meta
Meta's Segment Anything model. Track any subject as a mask that follows it.