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

# Cartesia Sonic 3 TTS

> Multilingual text-to-speech with 42 languages, voice cloning, and low latency streaming

<Card title="🎯 Cartesia Sonic 3: Multilingual Excellence" icon="bullseye">
  High-quality, low-latency TTS with support for 42 languages, voice cloning from \~5 second samples, and WebSocket streaming. Perfect for global deployments and multilingual voice agents.
</Card>

## Quick Setup

<Steps>
  <Step title="Get API Key">
    1. Visit [Cartesia.ai](https://cartesia.ai) and create an account
    2. Navigate to your API settings
    3. Copy your API key
  </Step>

  <Step title="Configure in Burki">
    1. Go to **AI Configuration** → **TTS** tab
    2. Select **Cartesia Sonic 3** as provider
    3. Paste your API key in the **TTS API Key** field
  </Step>

  <Step title="Choose Voice & Model">
    Select your preferred model and voice from the dropdowns
  </Step>
</Steps>

<Callout type="info">
  **Pricing**: Cartesia charges approximately \$0.015 per 1,000 characters. Check their website for current pricing details.
</Callout>

## Available Models

<CardGroup cols={2}>
  <Card title="⚡ sonic-3" icon="bolt">
    **Latest Model (Auto-updates)**

    Always uses the newest Sonic 3 improvements

    **Languages**: 42 languages
    **Best for**: Most applications, staying current
  </Card>

  <Card title="📌 sonic-3-2025-10-27" icon="thumbtack">
    **Stable Snapshot**

    Fixed version for consistent behavior

    **Languages**: 42 languages
    **Best for**: Production stability, testing
  </Card>
</CardGroup>

<Callout type="tip">
  **Recommendation**: Use **sonic-3** for development and most use cases. Use the dated snapshot for production if you need guaranteed consistency.
</Callout>

## Available Voices

### Preset Voices

<Accordion title="Female Voices">
  <CardGroup cols={3}>
    <Card title="Katie" icon="user">
      **Stable & Realistic**

      Perfect for voice agents

      `Voice ID: f786b574-daa5-4673-aa0c-cbe3e8534c02`
    </Card>

    <Card title="Tessa" icon="user">
      **Emotive & Expressive**

      Great for engaging conversations

      `Voice ID: 6ccbfb76-1fc6-48f7-b71d-91ac6298247b`
    </Card>

    <Card title="Sarah" icon="user">
      **Clear & Professional**

      Ideal for business applications

      `Voice ID: a0e99841-438c-4a64-b679-ae501e7d6091`
    </Card>
  </CardGroup>
</Accordion>

<Accordion title="Male Voices">
  <CardGroup cols={2}>
    <Card title="Kiefer" icon="user">
      **Stable & Realistic**

      Perfect for voice agents

      `Voice ID: 228fca29-3a0a-435c-8728-5cb483251068`
    </Card>

    <Card title="Kyle" icon="user">
      **Emotive & Expressive**

      Great for dynamic interactions

      `Voice ID: c961b81c-a935-4c17-bfb3-ba2239de8c2f`
    </Card>
  </CardGroup>
</Accordion>

<Accordion title="All Available Voices">
  | Voice      | Gender | Style               | Voice ID                               | Best For     |
  | ---------- | ------ | ------------------- | -------------------------------------- | ------------ |
  | **Katie**  | Female | Stable, Realistic   | `f786b574-daa5-4673-aa0c-cbe3e8534c02` | Voice agents |
  | **Kiefer** | Male   | Stable, Realistic   | `228fca29-3a0a-435c-8728-5cb483251068` | Voice agents |
  | **Tessa**  | Female | Emotive, Expressive | `6ccbfb76-1fc6-48f7-b71d-91ac6298247b` | Characters   |
  | **Kyle**   | Male   | Emotive, Expressive | `c961b81c-a935-4c17-bfb3-ba2239de8c2f` | Characters   |
  | **Sarah**  | Female | Clear, Professional | `a0e99841-438c-4a64-b679-ae501e7d6091` | Business     |
</Accordion>

## Language Support

<Callout type="info">
  **42 Languages Supported**: Cartesia Sonic 3 offers extensive multilingual support, making it ideal for global deployments.
</Callout>

<Accordion title="Supported Languages">
  <CardGroup cols={4}>
    <Card title="🇺🇸 English" icon="flag">`en`</Card>
    <Card title="🇪🇸 Spanish" icon="flag">`es`</Card>
    <Card title="🇫🇷 French" icon="flag">`fr`</Card>
    <Card title="🇩🇪 German" icon="flag">`de`</Card>
    <Card title="🇮🇹 Italian" icon="flag">`it`</Card>
    <Card title="🇵🇹 Portuguese" icon="flag">`pt`</Card>
    <Card title="🇨🇳 Chinese" icon="flag">`zh`</Card>
    <Card title="🇯🇵 Japanese" icon="flag">`ja`</Card>
    <Card title="🇰🇷 Korean" icon="flag">`ko`</Card>
    <Card title="🇮🇳 Hindi" icon="flag">`hi`</Card>
    <Card title="🇸🇦 Arabic" icon="flag">`ar`</Card>
    <Card title="🇷🇺 Russian" icon="flag">`ru`</Card>
  </CardGroup>

  **Additional Languages**: Dutch (`nl`), Polish (`pl`), Swedish (`sv`), Turkish (`tr`), Tagalog (`tl`), Bulgarian (`bg`), Romanian (`ro`), Czech (`cs`), Greek (`el`), Finnish (`fi`), Croatian (`hr`), Malay (`ms`), Slovak (`sk`), Danish (`da`), Tamil (`ta`), Ukrainian (`uk`), Hungarian (`hu`), Norwegian (`no`), Vietnamese (`vi`), Bengali (`bn`), Thai (`th`), Hebrew (`he`), Georgian (`ka`), Indonesian (`id`), Telugu (`te`), Gujarati (`gu`), Kannada (`kn`), Malayalam (`ml`), Marathi (`mr`), Punjabi (`pa`)
</Accordion>

## Voice Cloning

Cartesia supports creating custom voices from audio samples.

<Callout type="info">
  **Quick Cloning**: Create a custom voice from just \~5 seconds of clean audio.
</Callout>

### Requirements

| Requirement        | Value                                |
| ------------------ | ------------------------------------ |
| **Audio Duration** | \~5 seconds (recommended)            |
| **Formats**        | MP3, WAV, FLAC, M4A, OGG             |
| **Max File Size**  | 10MB                                 |
| **Sample Rate**    | 16kHz minimum recommended            |
| **Channels**       | Mono preferred                       |
| **Quality**        | Clean audio without background noise |

### Clone a Voice

<Steps>
  <Step title="Prepare Audio Sample">
    Record or select \~5 seconds of clear speech from the target voice
  </Step>

  <Step title="Upload via UI">
    1. Go to **Voice Configuration** in assistant settings
    2. Select **Cartesia** as provider
    3. Click **Clone Voice** button
    4. Upload your audio sample
  </Step>

  <Step title="Use Cloned Voice">
    The cloned voice appears in your voice dropdown and can be selected for your assistant
  </Step>
</Steps>

### Voice Cloning Tips

<Cards>
  <Card title="🎙️ Audio Quality" icon="microphone">
    Use clean audio without background noise - any noise will be reproduced in the clone
  </Card>

  <Card title="⏱️ Duration" icon="clock">
    \~5 seconds is ideal. Too short = poor quality, too long = unnecessary
  </Card>

  <Card title="👤 Single Speaker" icon="user">
    Audio should contain only one person speaking
  </Card>

  <Card title="🗣️ Natural Speech" icon="comments">
    Use conversational tone, not reading or acting
  </Card>
</Cards>

## Configuration Options

### Context Continuation

Cartesia maintains natural prosody across sentences using context continuation:

```python theme={null}
# First sentence starts new context
message = {
    "context_id": "ctx_0",
    "continue": False,  # New context
    "transcript": "Hello, welcome to our service."
}

# Subsequent sentences continue the context
message = {
    "context_id": "ctx_0", 
    "continue": True,  # Continue same context
    "transcript": "How can I help you today?"
}
```

### Flush Tags

Force immediate speech mid-sentence using the `<flush/>` tag:

```python theme={null}
# Text with flush tag
text = "Please hold<flush/> while I check your account."
# "Please hold" speaks immediately, then "while I check..." follows
```

### Audio Format

Cartesia automatically outputs the appropriate format for your telephony provider:

* **Twilio/Telnyx**: PCM μ-law @ 8kHz
* **Vonage**: PCM 16-bit @ 16kHz

## API Integration

<CodeGroup>
  ```python Python theme={null}
  from app.services.tts.tts_cartesia import CartesiaTTSService

  # Create Cartesia TTS instance
  tts = CartesiaTTSService(
      call_sid="unique_call_id",
      api_key="your_cartesia_api_key",
      voice_id="f786b574-daa5-4673-aa0c-cbe3e8534c02",  # Katie
      model_id="sonic-3",
      language="en"
  )

  # Start session
  await tts.start_session(audio_callback=your_callback)

  # Process text
  await tts.process_text("Hello, this is a test.")

  # End session
  await tts.end_session()
  ```

  ```javascript JavaScript theme={null}
  const response = await fetch("wss://api.cartesia.ai/tts/websocket", {
    headers: {
      "Authorization": `Bearer ${apiKey}`,
    }
  });

  // Send TTS request
  const message = {
    model_id: "sonic-3",
    transcript: "Hello, this is a test.",
    voice: {
      mode: "id",
      id: "f786b574-daa5-4673-aa0c-cbe3e8534c02"
    },
    language: "en",
    output_format: {
      container: "raw",
      encoding: "pcm_mulaw",
      sample_rate: 8000
    }
  };

  ws.send(JSON.stringify(message));
  ```

  ```curl cURL theme={null}
  # Clone a voice
  curl -X POST "https://api.cartesia.ai/voices/clone" \
    -H "Authorization: Bearer your_api_key" \
    -H "Cartesia-Version: 2025-04-16" \
    -F "clip=@voice_sample.mp3" \
    -F "name=My Custom Voice" \
    -F "language=en"
  ```
</CodeGroup>

## Performance Comparison

| Feature           | Cartesia Sonic 3            | ElevenLabs                | Deepgram Aura                    |
| ----------------- | --------------------------- | ------------------------- | -------------------------------- |
| **Latency**       | Vendor-reported low latency | Provider/config dependent | Vendor-reported very low latency |
| **Languages**     | 42                          | 70+                       | English + Spanish                |
| **Voice Quality** | High                        | Premium                   | Good                             |
| **Voice Cloning** | Yes (\~5 sec)               | Yes (1-25 min)            | No                               |
| **Best For**      | Multilingual agents         | Premium quality           | Speed-focused                    |

## Common Issues & Solutions

<Accordion title="Connection Issues">
  **Problem**: WebSocket connection fails

  **Solutions**:

  * Verify your `CARTESIA_API_KEY` is set correctly
  * Check network connectivity to api.cartesia.ai
  * Ensure API key has sufficient credits
</Accordion>

<Accordion title="Voice Not Found">
  **Problem**: Selected voice doesn't work

  **Solutions**:

  * Verify you're using a valid voice ID (UUID format)
  * If using a custom cloned voice, ensure it exists in your Cartesia account
  * Try one of the preset voices to test connectivity
</Accordion>

<Accordion title="Audio Quality Issues">
  **Problem**: Output sounds distorted or unclear

  **Solutions**:

  * Cartesia outputs 8kHz μ-law for Twilio compatibility
  * For higher quality, ensure you're using the correct output format for your use case
  * Check that your telephony provider supports the audio format
</Accordion>

## See Also

<CardGroup cols={3}>
  <Card title="⚡ Need Speed?" href="/tts-providers/deepgram">
    **Deepgram Aura** - speed-focused TTS for real-time calls
  </Card>

  <Card title="🎭 Premium Quality?" href="/tts-providers/elevenlabs">
    **ElevenLabs** - Industry-leading voice quality with 70+ languages
  </Card>
</CardGroup>

<Card title="🔗 Additional Resources" icon="link">
  **Official Documentation**: [Cartesia Docs](https://docs.cartesia.ai)

  **Voice Cloning**: [Clone API Reference](https://docs.cartesia.ai/api-reference/voices/clone)

  **WebSocket API**: [TTS WebSocket Guide](https://docs.cartesia.ai/api-reference/tts/websocket)
</Card>

***

<Card title="🚀 Ready to Use Cartesia?" icon="rocket">
  Head back to your [assistant configuration](/ai-configuration) and set up Cartesia Sonic 3 for multilingual voice experiences!
</Card>