Recipes
This page is an index of quickstarts, guides, and short recipes focused on particular use cases.
What's new
Vision agent using VoicePipelineAgent
Make outgoing calls with an agent
Using Twilio Voice
Agents
The following sections include guides, quickstarts, and code snippets that can help speed up the development process to creating your first, working agent.
Quickstarts and guides
- Voice agent with STT, LLM, TTS
Create an AI voice assistant with a pipeline of STT, LLM, and TTS models.
- Speech-to-speech agent
Create a speech-to-speech agent using the OpenAI Realtime API.
- An agent that answers incoming calls
Create an AI voice agent that responds to incoming phone calls using LiveKit Agents and LiveKit SIP.
- Make outbound calls with an agent
Create an AI voice agent that makes outbound calls for you. for you.
Examples and demos
- LiveKit Vision Demo
This LiveKit example project demonstrates how to build a sophisticated AI agent that can process realtime video feed, similar to the ChatGPT advanced voice with video feature.
- LivePaint
This LiveKit example project is a realtime drawing game where players compete to complete a drawing prompt as fast as possible, while being judged by a realtime AI agent that oversees the whole game.
- Speech-to-text example
Realtime transcription from voice to text.
LiveKit SIP
- Using Twilio Voice
How to use TwiML to accept incoming calls and bridge Twilio conferencing to LiveKit via SIP.
- Enhanced noice cancellation
Learn how to enable enhanced background noise cancellation audio filters for your LiveKit cloud project.
- HD voice with SIP
LiveKit SIP supports high fidelity calls by enabling HD Voice.