Eleven Labs API integration

2025-11-26 17:51 UTC gpt-5-1-thinking Open in ChatGPT ↗

Yes – ElevenLabs’ text-to-speech absolutely has an API, and it’s designed for exactly that kind of automation.

At a high level, you can:

Send a script (plain text) to their TTS endpoint
Choose a specific voice + model
Get back raw audio bytes (e.g. MP3)
Save those bytes as an audio file in your system

How it works (conceptually)

Create an API key

In the ElevenLabs dashboard you generate an API key, which you then send in the xi-api-key header. citeturn2view0
Pick a voice

Each voice has a voice_id. You either use one of their stock voices or a cloned/custom voice, and you’ll pass that ID in the URL. citeturn3view0

Call the Text-to-Speech endpoint

The core REST endpoint for converting text to speech looks like:

POST https://api.elevenlabs.io/v1/text-to-speech/{voice_id}

This converts your text into speech using the chosen voice and returns audio. citeturn4search1turn4search10

Typical headers and JSON body:

Headers:
  xi-api-key: YOUR_API_KEY
  Content-Type: application/json
  Accept: audio/mpeg   # or another supported format

JSON body:
{
  "text": "Your script goes here...",
  "model_id": "eleven_multilingual_v2",   // or another model
  "voice_settings": {
    "stability": 0.5,
    "similarity_boost": 0.5
  }
}

Save the returned audio as a file

The response body is the binary audio (e.g. MP3). You just write it to disk on your side—this is where your automation would loop over scripts, call the API, and save each result.

Example in Python (minimal)

Here’s a small example using requests to send a script and save an MP3:

import requests

API_KEY = "YOUR_API_KEY"
VOICE_ID = "21m00Tcm4TlvDq8ikWAM"  # example voice
URL = f"https://api.elevenlabs.io/v1/text-to-speech/{VOICE_ID}"

headers = {
    "xi-api-key": API_KEY,
    "Content-Type": "application/json",
    "accept": "audio/mpeg",
}

data = {
    "text": "This is a test script generated automatically.",
    "model_id": "eleven_multilingual_v2",
}

resp = requests.post(URL, headers=headers, json=data)
resp.raise_for_status()

with open("output.mp3", "wb") as f:
    f.write(resp.content)

They also provide official Python/TypeScript SDKs (e.g. elevenlabs.text_to_speech.convert(...)) that wrap this same functionality and can directly return audio you can save or stream. citeturn2view0turn0search17

So for your use case

✅ Automatically send scripts – yes, you can POST each script programmatically.
✅ Have it create audio files – the API returns the audio bytes; you save them as .mp3/.wav etc.
✅ Fully automatable pipeline – you can integrate this into any backend, batch process, or workflow tool.

If you tell me what stack you’re using (Python, Node, n8n, etc.), I can sketch the exact integration flow tailored to that.