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

# Voice Library & Cloning

> Choose from hundreds of built-in voices or clone your own. Create brand-consistent AI agents that sound exactly how you want.

## Voice Options Overview

TopCalls offers different voice options depending on the conversation mode you choose:

| Mode         | Voice Options               | Voice Cloning | Languages   | Latency     |
| ------------ | --------------------------- | ------------- | ----------- | ----------- |
| **Realtime** | Preset voices               | No            | Auto-detect | \~200-500ms |
| **Legacy**   | Hundreds of built-in voices | Yes           | 36+         | \~300-600ms |

<Note>
  Check available voices via the API:

  * All built-in voices: `GET /v1/voices/builtin`
  * Your account's voices (including cloned): `GET /v1/voices`

  Voice availability may change over time. Use the API endpoints for the most current list.
</Note>

## Realtime Mode Voices

In Realtime Mode, choose from preset voices optimized for low-latency conversations. The default is `alloy`.

```json theme={null}
{
  "mode": "realtime",
  "voice": "alloy"
}
```

<Note>
  Realtime Mode voices are optimized for ultra-low latency and natural conversation. Check available options via `GET /v1/voices/builtin`.
</Note>

## Legacy Mode: Full Voice Library

Legacy Mode gives you access to hundreds of voices across different styles, tones, and languages.

### Using Built-in Voices

Browse available voices via `GET /v1/voices/builtin` and use the voice ID in your calls:

```json theme={null}
{
  "mode": "legacy",
  "voice": "voice-id-from-api"
}
```

### Voice Cloning

Clone any voice with a short audio sample through the TopCalls dashboard:

**Step 1: Prepare Audio Samples**

* 1-5 audio files (MP3, WAV)
* At least 1 minute total duration
* Clear, high-quality recordings
* Single speaker, minimal background noise

**Step 2: Clone via Dashboard**

Upload your audio samples through the TopCalls dashboard. The platform processes them and creates a custom voice for your account.

**Step 3: Use Cloned Voice**

Once cloned, use the voice ID in your API calls:

```json theme={null}
{
  "mode": "legacy",
  "voice": "your_cloned_voice_id"
}
```

Your cloned voices appear in the `GET /v1/voices` endpoint alongside built-in voices.

<Warning>
  Voice cloning requires high-quality audio samples. Poor quality samples produce poor voice quality. Use professional recordings when possible.
</Warning>

## Choosing the Right Voice

### For Customer Support

Choose warm, professional voices that sound patient and helpful. Check available voices via the API and test a few to find the best fit.

### For Sales & Outreach

Choose confident, energetic voices that sound engaging. Match the voice to your brand personality.

### For Brand Consistency

Clone your brand spokesperson's voice and use the same voice across all channels.

## Multi-Language Voices

### Legacy Mode with Language Control

```json theme={null}
{
  "mode": "legacy",
  "stt_language": "es-ES",
  "voice": "spanish-voice-id"
}
```

Legacy Mode supports 36+ languages with proper dialect control. Set `stt_language` for accurate speech recognition and choose a voice that matches the target language.

Supported languages include English, Spanish, German, French, Dutch, Italian, Japanese, and many more.

## Best Practices

### Do This

* **Test voices**: Try different voices to find the best fit for your use case
* **Match tone**: Choose voices that match your brand personality
* **Consider the use case**: Support calls need different voices than sales calls
* **Use cloning for brands**: Clone spokesperson voices for consistency
* **Test quality**: Always test cloned voices before production

### Avoid This

* **Ignore latency**: Realtime Mode is faster but has fewer voice options
* **Poor audio samples**: Use high-quality recordings for cloning
* **Mismatched languages**: Make sure the voice language matches your instructions
* **Too many voices**: Stick to 1-2 voices for consistency

## Next Steps

<CardGroup cols={2}>
  <Card title="AI & Voice Customization" icon="microphone" href="/concepts/ai-voice">
    Control your AI's personality and behavior.
  </Card>

  <Card title="Making Calls" icon="phone" href="/guides/making-calls">
    Use custom voices when making calls.
  </Card>
</CardGroup>