Text to speech and speech to text for LiveKit Agents
The Cartesia plugin gives a LiveKit agent its voice and its ears. Use it with your own Cartesia API key, or call the same models through LiveKit Inference with no Cartesia account.
What you can do
Speak with Sonic
Sonic streams speech with under 90ms to first audio, in more than 40 languages, from the same plugin in Python or Node.
Transcribe with Ink
Use Ink as the agent's speech to text, in the Python plugin or through LiveKit Inference.
Choose who bills you
With your own API key, every voice in your Cartesia account works, including clones. With LiveKit Inference, LiveKit bills you and you skip the Cartesia account.
Get started
- Create a Cartesia API key, or skip this step if you use LiveKit Inference.
- Install the Cartesia plugin for LiveKit Agents.
- Set Cartesia as the text to speech and speech to text in your agent session, then choose a voice.
More integrations
Get started today
Talk to an expert.
Connect with a member of our team and learn how Cartesia can help you build world-class voice experiences.
Start building.
Access our models via API and bring a voice agent into production in minutes.