$0.05 / 1k chars
Single voice
TTS Turbo 2.5
Turn scripts, articles, product copy, and app text into natural narration with controls for voice, stability, similarity, speed, and language.
Text → audiox-api-keyup to 40k chars
ElevenLabs API · Muapi
Give every product a voice that feels alive.
Generate natural narration and expressive multi-speaker dialogue through one clean REST API. Bring your voice workflow to production with a single Muapi key, predictable pay-per-generation credits, and a shared async request pattern.
Live endpointsOne x-api-key headerPay per generation
api.muapi.aiLIVE
POST /api/v1/elevenlabs-tts-turbo-2-5
Text → audio
Request accepted200 OK
request_id: "req_8f2a..."
status: "processing"
Generated audio0:12
02
focused endpoints
40k
max TTS characters
$0.05
per 1k characters
01
API key to ship
Choose your workflow
Two endpoints. Every voice moment.
Start with fast single-voice narration or move to structured dialogue when your product needs more than one character.
$0.10 / 1k chars
Multiple speakers
Text to Dialogue V3
Compose expressive conversations from structured speaker turns, assigning a distinct voice to every character in the script.
Dialogue → audiox-api-keyvoice per turn
Made for real products
Audio that fits the moment.
Keep the integration small and the creative surface wide. Use the same request shape across narration, dialogue, localization, and interactive experiences.
Compare all text-to-speech modelsNarration
Ship voiceovers for product videos, courses, articles, and accessibility features.
Localization
Pass a language code and create natural audio for global products and content.
Fast iteration
Generate on demand with a compact JSON request and poll the result when ready.
Quick start
From prompt to playable audio in three calls.
The API is intentionally simple: submit JSON, keep the request ID, then poll for the generated audio URL.
01
Create a Muapi API key
Create a key in your workspace and keep it in an environment variable.
02
Submit to an endpoint
Choose TTS Turbo for one voice or Dialogue V3 for a structured conversation.
03
Poll for the result
Use the returned request ID until the task is complete, then use the audio URL in your app.
$0.05 / 1k chars · pay per generation
quick-start.sh
curl -X POST "https://api.muapi.ai/api/v1/elevenlabs-tts-turbo-2-5" \
-H "x-api-key: YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"prompt": "Your workspace is ready.",
"voice_id": "21m00Tcm4TlvDq8ikWAM",
"speed": 1.0
}'
# → {"request_id":"REQUEST_ID"}
curl "https://api.muapi.ai/api/v1/predictions/REQUEST_ID/result" \
-H "x-api-key: YOUR_API_KEY"Tune the delivery
A small schema with a lot of range.
Start with the defaults, then shape the performance that makes your product sound like itself.
voice_idPick a built-in voice or pass a compatible custom voice ID.
stabilityBalance consistency and expressive variation from 0 to 1.
similarity_boostTune how closely the result follows the selected voice.
speedSet the speaking rate from 0.7× to 1.2× for narration.
Built for voice products
Narration, dialogue, podcasts, games, and more.
Questions, answered
Everything you need to ship.
What is included in the ElevenLabs API page?
You get two focused endpoints: TTS Turbo 2.5 for fast single-voice narration and Text to Dialogue V3 for structured multi-speaker audio. Both use the same Muapi authentication and submit-then-poll pattern.
Can I use a custom voice ID?
Yes. Pass a compatible ElevenLabs voice_id in your request. The endpoint schema also includes a curated set of built-in voice options to make testing easy.
How does pricing work?
TTS Turbo 2.5 is listed from $0.05 per 1,000 characters, while Dialogue V3 is listed at $0.10 per 1,000 characters. Usage is pay per generation with no monthly subscription required.
How do I get the generated audio?
The submit response returns a request_id. Poll the predictions result endpoint until status is completed, then read the generated audio URL from the response.
Your next feature has a voice