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

<AgentInstructions>

## Submitting Feedback

If you encounter incorrect, outdated, or confusing documentation on this page, submit feedback:

POST https://docs.poly.ai/feedback

```json
{
  "path": "/voice/agent",
  "feedback": "Description of the issue"
}
```

Only submit feedback when you have something specific and actionable to report.

</AgentInstructions>

# Agent Voice

> Configure your agent's voice settings.

Use this page to select and fine-tune the voice your callers hear. The voice you choose shapes how callers perceive your brand – a mismatch in accent, tone, or quality can undermine an otherwise well-built agent.

Configure voice settings in **Channels > Voice > Agent Voice**.

## Voice sections

The page displays two voice configurations:

| Section        | Purpose                                                 |
| -------------- | ------------------------------------------------------- |
| **Agent**      | The main voice your users interact with during calls    |
| **Disclaimer** | A separate voice for legal or informational disclaimers |

## Selecting a voice

1. Click **Change** next to either the Agent or Disclaimer section.
2. The **[Voice library](/voice/voice-library)** opens.
3. Browse voices using the **Explore** tab or access saved voices in **Favorites**.
4. Filter by **language**, **region**, and **gender**.
5. Preview voices with custom text before selecting.
6. Click **Select** to apply the voice.

Use the Voice library's preview feature with text that matches your agent's typical responses to ensure the voice fits your use case.

## Voice settings

Fine-tune the selected voice's parameters:

1. Click the **Settings** icon <Icon icon="gear" iconType="solid" /> next to the voice.
2. Adjust the sliders in the Voice Settings popup:

<ParamField path="Stability" type="slider">
  Controls how consistent the voice sounds over time. Higher values keep tone and delivery more uniform across responses.

  **Recommended:** Set to 0.7 or above for reliable performance.
</ParamField>

<ParamField path="Clarity and Similarity" type="slider">
  Adjusts how closely the generated voice matches the original recording. Higher values sound more realistic and detailed, but may be less resilient to unexpected prompts. Lower values smooth the voice slightly, making it better suited for dynamic or ASR-driven content.
</ParamField>

<ParamField path="Style" type="slider">
  Controls stylistic expressiveness of the voice (0.0–1.0). Higher values produce more expressive speech with varied intonation. Lower values keep delivery more neutral and consistent. Only supported by certain TTS providers and voice models.
</ParamField>

3. Click **Done** to save changes.

## Publishing changes

After updating voice settings:

1. Review your configuration on the Agent Voice page.
2. Click **Publish** in the top right corner to apply changes to your live agent.

<Warning>
  Voice changes only take effect after publishing. Test your agent after publishing to confirm the new voice sounds as expected.
</Warning>

## Disclaimer voice configuration

If disclaimer audio is configured separately, the page provides a link to **Channels > Voice > Voice Configuration** for those settings.

The Disclaimer voice is optional. If not configured, the Agent voice handles all agent speech.

## Related pages

<CardGroup cols={2}>
  <Card title="Voice library" icon="list" href="/voice/voice-library">
    Browse and select voices
  </Card>

  <Card title="Choosing a good voice" icon="ear-listen" href="/voice/choosing-a-good-voice">
    Guidelines for voice selection and matching
  </Card>

  <Card title="Voice configuration" icon="gear" href="/voice/voice-configuration">
    Configure greeting and disclaimer settings
  </Card>

  <Card title="Add a voice" icon="plus" href="/voice/add-a-new-voice">
    Configure custom voices programmatically
  </Card>
</CardGroup>