> ## Documentation Index
> Fetch the complete documentation index at: https://docs.getnetra.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# LiveKit

> Connect LiveKit voice agents to Netra for evaluation and simulation. Auto-instrument your agent, configure credentials in the dashboard, and run voice simulations.

<img src="https://mintcdn.com/netra/I4AMRwYHDuF5FbHN/images/integration-logos/voice-frameworks/livekit.png?fit=max&auto=format&n=I4AMRwYHDuF5FbHN&q=85&s=1857c904bf8826b18d78c2ad27450068" alt="LiveKit" width="332" height="80" data-path="images/integration-logos/voice-frameworks/livekit.png" />

LiveKit is an open-source real-time communication platform for building voice and video applications. Netra integrates with LiveKit to evaluate voice agents through automated simulations—placing simulated calls, recording conversations, and scoring quality.

## How It Works

Netra connects to your LiveKit project to run voice simulations:

1. You configure your LiveKit credentials in the Netra dashboard
2. Netra creates a room and dispatches your agent into it
3. A simulated caller joins the room and speaks with your agent
4. The conversation is recorded, transcribed, and evaluated

## Configuration

Register your LiveKit agent under **Evaluation → Agents** in the Netra dashboard. Select **Voice** as the modality and **LiveKit** as the platform.

| Field          | Description                                      |
| -------------- | ------------------------------------------------ |
| **Server URL** | WebSocket URL (wss\://) for your LiveKit project |
| **API Key**    | LiveKit API key for authentication               |
| **API Secret** | LiveKit API secret for token generation          |
| **Agent Name** | The name of your agent or room to connect to     |

You can find these values in your [LiveKit Cloud dashboard](https://cloud.livekit.io) under **Settings → API Keys**.

## SDK Integration

Instrument your LiveKit agent with the Netra SDK to enable OTLP span emission. This provides visibility into internal processing steps like LLM calls, TTS generation, and tool usage.

<CodeGroup>
  ```python Python theme={null}
  from netra import Netra

  Netra.init(app_name="my-livekit-agent")
  ```

  ```typescript TypeScript theme={null}
  import { Netra } from "netra-sdk";

  await Netra.init({ appName: "my-livekit-agent" });
  ```
</CodeGroup>

## Supported Features

| Feature                         | Description                                        |
| ------------------------------- | -------------------------------------------------- |
| **Room-based communication**    | Standard LiveKit room protocol with agent dispatch |
| **Token authentication**        | Netra mints room-scoped tokens for secure access   |
| **Background noise simulation** | Inject realistic background audio during calls     |
| **Real-time transcription**     | Live transcript as the conversation progresses     |
| **Stereo recording**            | Full session audio available for playback          |
| **SDK tracing**                 | OTLP spans from your agent for deep observability  |

## Related

* [Voice Simulation Agents](/Simulations/voice-simulations/agents) — Register your LiveKit agent in Netra
* [Voice Simulation Datasets](/Simulations/voice-simulations/Simulations) — Create test scenarios for your voice agent
* [Vapi Integration](/Integrations/voice-frameworks/Vapi) — Compare with Vapi integration