Skip to content

Mage API · Audio model

Eleven Speech API

ElevenLabs v4: lifelike speech you can direct

POST https://api.mage.space/v1/eleven_speech/generate

At a glance

The Eleven Speech API runs Eleven v4 through Mage at POST https://api.mage.space/v1/eleven_speech/generate. A request at the default settings costs 30 Gems (about $0.03), with no subscription. Eleven Speech takes no @handle mentions.

  • AudioOutput
  • 1Variant
  • 30Gems per clip, default
  • $0.03About, in USD

Overview

What the Eleven Speech API does

ElevenLabs v4 text to speech in 16 voices and 90+ languages, directed line by line with audio tags like [whispers] or [laughs].

ElevenLabs' expressive text to speech in 90+ languages, as an MP3 as long as the text. Audio tags in square brackets direct the delivery and add sound cues: [whispers], [laughs], [sarcastic], [door slams].

It generates from the prompt alone.

Quick start

Call the Eleven Speech API

Set MAGE_API_KEY, submit to /eleven_speech/generate, and poll status_url until the request completes.

cURL
STATUS_URL=$(curl --request POST 'https://api.mage.space/v1/eleven_speech/generate' \
  --header "Authorization: Bearer $MAGE_API_KEY" \
  --header 'Content-Type: application/json' \
  --data '{
  "prompt": "Every lighthouse keeps one promise... to stay lit until morning. [softly] And this one never broke it.",
  "model_id": "eleven-v4"
}' | jq -r '.status_url')

# Repeat until status is "completed"; result.url is the file
curl "$STATUS_URL" --header "Authorization: Bearer $MAGE_API_KEY"
TypeScript
// npm install @mage-space/sdk
import Mage from '@mage-space/sdk';

const mage = new Mage(); // reads MAGE_API_KEY

const request = await mage.run('eleven_speech', {
  prompt: 'Every lighthouse keeps one promise... to stay lit until morning. [softly] And this one never broke it.',
  model_id: 'eleven-v4',
});

console.log(request.result.url);
Python
# pip install mage-space
from mage_space import Mage

mage = Mage()  # reads MAGE_API_KEY

request = mage.run("eleven_speech", {
    "prompt": "Every lighthouse keeps one promise... to stay lit until morning. [softly] And this one never broke it.",
    "model_id": "eleven-v4",
})

print(request["result"]["url"])

Models

Eleven Speech model_id values

Send model_id to pick a variant. A request without one runs the default.

model_idNameDefault
eleven-v4Eleven v4Yes

Options

Eleven Speech parameters

Allowed values across all variants; a variant may offer fewer. The docs list the options per variant.

FieldValues
voicedarian, talia, elara, baxter, eldrin, elowen, kaelen, lawrence, alicia, maisie, warren, jade, caleb, sawyer, florence, wyatt

Rules

What the Eleven Speech API enforces

  • The prompt is the text to speak, at most 5000 characters, audio tags included.
  • `voice` is one of darian, talia, elara, baxter, eldrin, elowen, kaelen, lawrence, alicia, maisie, warren, jade, caleb, sawyer, florence, wyatt.
  • `stability` runs from 0 (expressive) to 1 (steady).

More audio models

Other audio generation APIs on Mage

  • Audio · from 19 Gems

    Seed Audio API

    Voices, music, and sound effects from one prompt

  • Audio · from 113 Gems

    Eleven Music API

    ElevenLabs songs, with vocals or without

FAQ

Frequently asked questions

How much does the Eleven Speech API cost?
A Eleven Speech request at the default settings (Eleven v4) costs 30 Gems, about $0.03. Other resolutions, durations, and variants change the price; the response reports the exact charge. A generation that fails is refunded; one Mage's content policy blocks is not.
What is the Eleven Speech API endpoint?
POST https://api.mage.space/v1/eleven_speech/generate with an Authorization: Bearer header carrying your Mage API key. The response includes a status_url to poll for the result.
Which Eleven Speech models can I call?
eleven-v4 (Eleven v4). The default is Eleven v4.
Does Eleven Speech support reference images or characters?
Eleven Speech takes no image, video, or audio inputs. Eleven Speech takes no @handle mentions.
Can I use Eleven Speech from Claude or ChatGPT?
Yes. Connect the Mage MCP server to Claude, ChatGPT, Claude Code, Grok, or Cursor and ask for Eleven Speech by name. Your assistant reads its options with get_model and quotes the price before it generates.
How long are Eleven Speech results kept?
Results and inputs are kept for 30 days. Download anything you want to keep to your own storage.

Start building with Mage

One account, one Gems balance, every model. Pay only for what you generate.