Grok Text-to-Speech
Turn text into natural, expressive speech. Grok Text-to-Speech is xAI's voice synthesis model on Replicate. Input text to get lifelike audio. It offers 5 unique voices, inline speech tags for delivery control and multiple audio formats.
Generate your speech
Enter your text, pick a voice, and generate natural speech in seconds.
Everything you need for voice generation
Grok Text-to-Speech combines natural prosody, rich controls, and flexible outputs — built for content creators, developers, and businesses.
5 Distinct Voices
eve, ara, rex, sal, leo — each with its own personality and tone, from energetic to authoritative.
17+ Languages
Arabic, Bengali, Chinese, English, French, German, Hindi, Indonesian, Italian, Japanese, Korean, Portuguese, Russian, Spanish, Turkish, Vietnamese with automatic language detection.
Inline Speech Tags
Control pacing, emotion and delivery with [pause], [laugh], <whisper>, <slow>, <soft> and more.
Multiple Formats
MP3 (default), WAV, raw PCM, μ-law, A-law — from high-fidelity to telephony-grade.
Configurable Quality
Sample rates 8 kHz – 48 kHz, MP3 bit rates 32 – 192 kbps. Studio or telephony, your choice.
Long Inputs
Up to 15,000 characters per request with built-in text normalization for numbers, currencies and abbreviations.
Five distinct voices
Each voice has its own personality and tone — pick the one that fits your content.
Eve
Energetic, upbeat
Default voice — engaging and enthusiastic.
Ara
Warm, friendly
Balanced and conversational.
Rex
Confident, clear
Professional and articulate.
Sal
Smooth, balanced
Versatile across content types.
Leo
Authoritative, strong
Commanding — great for instructional content.
Control pacing, emotion, and delivery
Add inline tags to control how the voice delivers your text. Inline tags drop in at a specific point; wrapping tags change how a span of text is delivered.
[pause]Insert a short pause[long-pause]Insert a longer pause[laugh]Add a laugh sound[sigh]Add a sigh[breath]Add a breath sound<whisper>...</whisper>Whispered delivery<slow>...</slow>Slower pacing<soft>...</soft>Softer, quieter deliverySo I walked in and [pause] there it was. [laugh] I could not believe it.
I need to tell you something. <whisper>It is a secret.</whisper> Pretty cool, right?
Output formats for every use case
Defaults are MP3 at 24 kHz / 128 kbps. For studio-grade audio, use MP3 or WAV at 44.1 kHz or 48 kHz with 192 kbps. For telephony, use μ-law or A-law at 8 kHz.
MP3
General use — wide compatibility, good compression.
WAV
Lossless audio — editing, post-production.
Raw PCM
Raw audio — real-time processing pipelines.
μ-law
Telephony (G.711 μ-law).
A-law
Telephony (G.711 A-law).
17+ supported languages
The model auto-detects the language by default. Set a specific language code to force a language or get more consistent results.
Frequently asked questions
What is Grok Text-to-Speech?
Grok Text-to-Speech is xAI's voice synthesis model. Send it text and get back natural, expressive speech audio. It runs on Replicate and is available through imagine-video.
How many voices are available?
There are 5 distinct voices — eve (energetic), ara (warm), rex (confident), sal (smooth), and leo (authoritative). Each has its own personality and tone.
What output formats are supported?
MP3 (default), WAV (lossless), raw PCM, μ-law, and A-law. Defaults are MP3 at 24 kHz / 128 kbps. For studio-grade audio use MP3 or WAV at 44.1 kHz or 48 kHz with 192 kbps.
What are speech tags?
Speech tags let you control delivery inline. Inline tags like [pause], [laugh], [sigh], [breath] drop in at a specific point. Wrapping tags like <whisper>, <slow>, <soft> change how a span of text is delivered.
What languages are supported?
Arabic, Bengali, Chinese (Simplified), English, French, German, Hindi, Indonesian, Italian, Japanese, Korean, Portuguese, Russian, Spanish, Turkish, Vietnamese. The model auto-detects language by default.
Is the generated audio stored?
Yes. Generated audio is uploaded to our R2 storage for permanent access. You can play it directly on the page or download it.
Ready to give your text a voice?
Try Grok Text-to-Speech on imagine-video. Generate natural, expressive speech in seconds.