Agents

Learn best practices for building multi-agent architectures including session state management, chat context handling, TaskGroup patterns, and dynamic per-client routing.

Understand the hard cap on agent deployments and how to build a multi-tenant agent that scales without provisioning more slots.

A technical comparison of LiveKit Agents and Pipecat for building voice AI applications, covering architecture, capabilities, integrations, and developer experience.

Learn how to protect sensitive user data by redacting PII from agent logs, transcripts, and observability data in LiveKit.

Learn the difference between NC and BVC, and how to select the right noise cancellation model based on your participant type.

A demo of Inworld's text-to-speech model being used with a LiveKit Agent to play an AI-driven tabletop style RPG in real time.

Learn how to build voice agents with emotional understanding using the Grok Voice API on LiveKit. This demo shows how xAI and LiveKit partnered to bring Grok's voice technology to developers.

Learn how to handle payments in LiveKit voice agents using a hybrid architecture that keeps sensitive payment data off the voice channel.

Optimize network proximity and key metrics like TTFT and TTFB to reduce latency in voice agents that use Speech-to-Text, Text-to-Speech, and Large Language Models.

Configure a LiveKit Agent to accept audio input via speech-to-text while responding only with text, and learn how to receive the text responses on your frontend.

Learn how to automatically end calls after periods of silence using the built-in user away timeout and custom retry mechanisms.

Learn how to accurately detect when a voice agent has finished speaking using SpeechHandle, and how to notify web clients.

Learn how to create a voice agent that repeats a greeting when there's no response, and gracefully ends the call after multiple attempts using the away state.

Learn how to configure your LiveKit agent to process audio from a specific participant in multi-participant rooms.

Learn how to create an evaluation agent that plays the role of your user, tests your agent, and provides feedback on the interaction.

With a few commands from the CLI, you can start talking to your LiveKit Agent directly through your console using LiveKit Inference's free tier.

A demo of Deepgram's Flux model being used with a LiveKit Agent to take medical notes and diagnose based on symptoms.

Learn how to build a production-ready AI voice agent with zero code using LiveKit Agent Builder and n8n. Create a restaurant voice agent that answers menu questions and takes orders—all without writing a single line of code.

Learn how to fix RtcError: InvalidState - failed to capture frame errors when working with LiveKit AudioSource by adjusting the queue_size_ms parameter.

Learn how to diagnose and resolve speech-to-text detection failures when STT fails to register user speech, including noise cancellation issues, STT provider limitations, and configuration recommendations.

Learn how to properly set up and test your agents using the LiveKit testing framework with both standard LLM models and realtime models.

Understand which VAD and turn detection parameters apply to your agent type, and how to configure them for optimal voice interaction.

In Agents 1.3+, job dispatch stops after the first rejection by default. Learn how to route jobs to specific agents using explicit dispatch instead.

Learn why LiveKit egress recordings of agent sessions have extra audio at the end, and how to prevent it by deleting the room or manually stopping egress when the human participant leaves.