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

# Voices

> Choose and configure the voice your AI Teammate speaks with.

Every AI Teammate speaks through a **Text-to-Speech (TTS)** engine that converts the LLM's text response into natural, expressive audio in real time. TruGen integrates with **ElevenLabs** and **Cartesia** to deliver high-quality, human-like voices across a wide range of tones, styles, and languages.

## How It Works

When the LLM generates a response, it is passed to the TTS engine which renders it as speech before being streamed back to the user through the avatar. The whole process happens in under a second.

The voice your agent uses is set via the `voice_id` field in your agent configuration. Every voice has a unique ID — copy it from the [Voice Library](/docs/agents/voices/library) and paste it directly into your agent.

## Setting a Voice

Set the `voice_id` when creating or updating an agent via API:

<CodeGroup>
  ```bash cURL theme={null}
  curl --request POST \
    --url https://api.trugen.ai/v1/ext/agent \
    --header 'Content-Type: application/json' \
    --header 'x-api-key: YOUR_API_KEY' \
    --data '{
      "agent_name": "Support Agent",
      "avatars": [
        {
          "config": {
            "llm": {
              "provider": "openai",
              "model": "gpt-4.1-mini"
            },
            "stt": { "provider": "deepgram", "model": "nova-3" },
            "tts": {
              "provider": "elevenlabs",
              "voice_id": "FGY2WhTYpPnrIDTdsKH5"
            }
          },
          "persona_name": "Support Agent",
          "persona_prompt": "You are a helpful support agent."
        }
      ]
    }'
  ```
</CodeGroup>

<Note>You can also set the voice directly from the **Create New Agent** form in the [Developer Portal](https://app.trugen.ai) without writing any code.</Note>

## Choosing the Right Voice

| Goal                          | Recommendation                                              |
| ----------------------------- | ----------------------------------------------------------- |
| Professional / corporate tone | Chris, Will, Evan                                           |
| Warm / conversational         | Rachel, Laura, Hope                                         |
| Multilingual coverage         | Any voice — ElevenLabs auto-adapts to the detected language |
| Regional accent               | Yatin (South Asian), Omar (Middle Eastern)                  |

## What's Next?

<CardGroup cols={2}>
  <Card title="Voice Library" icon="microphone" href="/docs/agents/voices/library">
    Browse all available voices and copy their IDs.
  </Card>

  <Card title="Providers" icon="cloud" href="/docs/providers/overview">
    See all STT, LLM, and TTS providers available on TruGen.
  </Card>
</CardGroup>