LiveKit Agents and Pipecat

Learn in five minutes

• LiveKit's pipeline page shows an AgentSession that takes speech-to-text, a language model, and text-to-speech.
• Pipecat's overview calls Pipecat an open source Python framework that orchestrates services, transports, and audio.
• Pipecat's pipeline page shows transport input, speech-to-text, a user context aggregator, the language model, text-to-speech, transport output, then an assistant aggregator.
• That page says processors push frames along, and it lists InterruptionFrame as a system frame.
• Reading either page is not a deployment you completed.

Both are orchestrators in their own docs. The model is still a box you configure. This lab does not install either framework.

This section HTML

Visual

Diagram SVGDiagram PNGDiagram HTML

Flow SVGFlow PNGFlow HTML

Two orchestrators on paper

1. Write LiveKit's three arguments: stt, llm, tts.
2. Write Pipecat's processor order from the pipeline page.
3. Circle the transport in Pipecat's list. That is the audio's door.
4. Do not run an install command in this room.

This section HTML

Quiz

This section HTML

Ecosystem

Orchestrator

The framework that orders the stages. It is not itself the language model.

Sisters: Speech-to-text, Language model, Text-to-speech, WebRTC, SIP.

A cascaded pipeline and a speech-to-speech model are two shapes the LiveKit docs name. The xAI Speech to Speech API is a hosted speech-to-speech session.

Pin the model id, the voice, and the date you read the docs.

Config

• Write the goal in one sentence.
• Name the input and the allowed output.
• Save the trace next to the result.

Troubleshooting

• The answer looks fluent and is wrong. Check whether a source passage actually contains the claim.
• The loop never stops. Set a step cap and stop when the tool result repeats.

This section HTML