SkillAgentSearch skills...

Elevenlabslib

Full python wrapper for the elevenlabs API.

Install / Use

/learn @lugia19/Elevenlabslib
About this skill

Quality Score

0/100

Supported Platforms

Universal

README

elevenlabslib

<a href='https://ko-fi.com/lugia19' target='_blank'><img height='35' style='border:0px;height:46px;' src='https://az743702.vo.msecnd.net/cdn/kofi3.png?v=0' border='0' alt='Buy Me a Coffee at ko-fi.com' /></a> PyPI - Downloads PyPI GitHub last commit

Python wrapper for the full elevenlabs API.

This project is now deprecated. The API simply changes too quickly and has become too broad for me to keep up with.

The main reason is the different approach to playback. By doing playback purely within python instead of piping to an external process, there are a couple of important extra things that can be done, such as:

  • Playback on a specific output device
  • Running functions exactly when the playback begins and ends
  • Controlling the playback from within python

Documentation now available at https://elevenlabslib.readthedocs.io/en/latest/

Installation

Just run pip install elevenlabslib, it's on pypi.

Note: On Linux, you may need to install portaudio. On debian and derivatives, it's sudo apt-get install libportaudio2, and possibly also sudo apt-get install python3-pyaudio.

IMPORTANT: The library requires libsndfile v1.1.0 or newer, as that is when mp3 support was introduced. This won't be an issue on Windows, but may be relevant on other platforms. Check the soundfile repo for more information.

Usage

For a far more comprehensive example, check example.py or the docs.

Here is a very simple usage sample.

  • Retrieves a voice based on the name
  • Plays back (using the included playback functions that use sounddevice) all its samples (and the preview)
  • Generates and plays back a new audio
  • Deletes the newly created audio from the user history
from elevenlabslib import *

user = User("API_KEY")
voice = user.get_voices_by_name_v2("Rachel")[0]  # This is a list because multiple voices can have the same name

voice.generate_play_audio_v2("Test.", playbackOptions=PlaybackOptions(runInBackground=False))

for historyItem in user.get_history_items_paginated():
    if historyItem.text == "Test.":
        # The first items are the newest, so we can stop as soon as we find one.
        historyItem.delete()
        break
View on GitHub
GitHub Stars158
CategoryDevelopment
Updated29d ago
Forks29

Languages

Python

Security Score

100/100

Audited on Mar 2, 2026

No findings