Grok TTS
by xAI
xAI's voice model for reads that perform.
Five voices, laughs on cue, words you can whisper.

Key features
Technical specifications
5
Built-in voices, each speaking every supported language
15,000 chars
Maximum script length per generation on Morphic
Inline + wrap
Sound cues inline, delivery tags around words
MP3, 24 kHz
128 kbps audio, ready for a timeline
Use cases

Explainers and tutorials
Rex for a clear walkthrough, Sal for a calmer one. Regenerate a single step when the product changes.

Ads and promos
Eve's energy for a launch read, with a whispered aside wrapped in a tag so only those words drop.

Localized voice-over
Keep the same voice across markets. Paste each translation and the voice follows the language.

Story and podcast narration
Long scripts in one pass, with pauses and laughs written where the beat lands.

Assistant and app voices
Ara's warmth for onboarding, help and spoken notifications that should sound friendly.

Announcements and trailers
Leo's weight for a trailer line or an announcement that needs to sound like it matters.
Prompt examples


Whispered secret
Don't tell anyone, but <whisper>the surprise is on Friday</whisper>.


Slow emphasis
Read the label. Then <slow>read it again</slow> before you plug it in.


Overview
Grok TTS is xAI's text-to-speech model, on Morphic since July 2026. It reads a script in one of five built-in voices, performs sounds like a laugh or a pause where you write them, and changes the delivery of any words you wrap in a tag.
What Grok TTS does differently
Grok gives you two kinds of direction inside the script. Sound cues sit inline in square brackets, so [laugh] happens at that point in the line. Delivery tags wrap words, so <whisper> changes only what sits between the opening and closing tag. The voices are few but not tied to a language: each of the five speaks every language Grok supports, which keeps one voice consistent from the English cut to every translation.
Grok TTS on Morphic
On Morphic, Grok TTS sits in the Audio model picker for Speech as xAI Grok TTS. Choose Eve, Ara, Rex, Sal or Leo, paste a script of up to 15,000 characters with any tags inline, and generate. Grok detects the language from the text, and the audio comes back as MP3 on your Canvas. For a larger voice library or cloned voices, ElevenLabs v4 runs in the same picker.
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.