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

# Audio

> Configure TTS (text-to-speech), STT (speech-to-text), and avatar engines.

<Info>Admin › Settings › Audio</Info>

Configure voice input/output engines.

<Frame caption="Audio settings tab full screen">
  <img src="https://mintcdn.com/cloocus/Or1RjU_AzOVJRxuS/images/admin/settings-audio-full.png?fit=max&auto=format&n=Or1RjU_AzOVJRxuS&q=85&s=aec0823e6774e334e7bf8349b0f2285f" alt="Audio settings tab full screen" width="2880" height="1800" data-path="images/admin/settings-audio-full.png" />
</Frame>

***

## TTS (Text-to-Speech)

Configure the engine that converts AI responses to speech.

| Setting                | Description                                  | Default       |
| ---------------------- | -------------------------------------------- | ------------- |
| **Engine**             | TTS service selection                        | `Web API`     |
| **Model**              | Per-engine TTS model                         | -             |
| **Voice**              | Voice type                                   | -             |
| **Response splitting** | TTS response split unit (sentence/paragraph) | `Punctuation` |

<Tabs>
  <Tab title="System (Web API)">
    Uses the browser's built-in Web Speech API. Ready to use without separate configuration.
  </Tab>

  <Tab title="Transformers (Local)">
    Synthesizes speech with the server's built-in Transformers model. Requires no external API key.
  </Tab>

  <Tab title="OpenAI">
    | Setting          | Description                                         |
    | ---------------- | --------------------------------------------------- |
    | **API Base URL** | OpenAI API Endpoint                                 |
    | **API Key**      | API authentication key                              |
    | **Model**        | e.g., `tts-1`, `tts-1-hd`                           |
    | **Voice**        | `alloy`, `echo`, `fable`, `onyx`, `nova`, `shimmer` |
  </Tab>

  <Tab title="Azure AI Speech">
    | Setting           | Description                         |
    | ----------------- | ----------------------------------- |
    | **API Key**       | Azure AI Speech Service key         |
    | **Region**        | Azure region (e.g., `koreacentral`) |
    | **Output Format** | Audio output format                 |
  </Tab>

  <Tab title="ElevenLabs">
    | Setting     | Description          |
    | ----------- | -------------------- |
    | **API Key** | ElevenLabs API key   |
    | **Voice**   | Voice selection      |
    | **Model**   | ElevenLabs TTS model |
  </Tab>

  <Tab title="Google Cloud TTS">
    | Setting                        | Description                           |
    | ------------------------------ | ------------------------------------- |
    | **Language Filter (Optional)** | Voice language filter (e.g., `ko-KR`) |
    | **Service Account Key**        | Global key or Custom JSON key         |
  </Tab>

  <Tab title="Gemini TTS">
    | Setting                 | Description                                                |
    | ----------------------- | ---------------------------------------------------------- |
    | **Location**            | Region (default: `us-central1`)                            |
    | **Service Account Key** | Global key or Custom JSON key                              |
    | **TTS Voice**           | Gemini TTS voice selection                                 |
    | **Model**               | Gemini TTS model (default: `gemini-2.5-flash-preview-tts`) |
  </Tab>
</Tabs>

***

## STT (Speech-to-Text)

Configure the engine that converts user voice input to text.

| Setting    | Description           | Default           |
| ---------- | --------------------- | ----------------- |
| **Engine** | STT service selection | `Whisper (local)` |
| **Model**  | Per-engine STT model  | -                 |

<Tabs>
  <Tab title="Whisper (Local)">
    Transcribes with the server's built-in faster-whisper model. **Default**, requires no external API key.
  </Tab>

  <Tab title="System (Web API)">
    Uses the browser's built-in Web Speech API.
  </Tab>

  <Tab title="OpenAI (Whisper)">
    | Setting          | Description             |
    | ---------------- | ----------------------- |
    | **API Base URL** | OpenAI API Endpoint     |
    | **API Key**      | API authentication key  |
    | **Model**        | Whisper model selection |
  </Tab>

  <Tab title="Azure AI Speech">
    | Setting     | Description                                                  |
    | ----------- | ------------------------------------------------------------ |
    | **API Key** | Azure AI Speech Service key                                  |
    | **Region**  | Azure region                                                 |
    | **Locales** | Recognition languages (comma-separated, e.g., `ko-KR,en-US`) |
  </Tab>

  <Tab title="Deepgram">
    | Setting     | Description      |
    | ----------- | ---------------- |
    | **API Key** | Deepgram API key |
  </Tab>

  <Tab title="Google Cloud Speech">
    | Setting                 | Description                             |
    | ----------------------- | --------------------------------------- |
    | **Project ID**          | Google Cloud project ID                 |
    | **Location**            | Region (default: `global`)              |
    | **Language Codes**      | Recognition languages (default: `auto`) |
    | **Service Account Key** | Global key or Custom JSON key           |
  </Tab>
</Tabs>

***

## Avatar

Configure avatars displayed during AI responses.

| Setting       | Description                       |
| ------------- | --------------------------------- |
| **Engine**    | Avatar service selection          |
| **API Key**   | Avatar service authentication key |
| **Region**    | Service region                    |
| **Character** | Avatar character selection        |
| **Style**     | Avatar style                      |
| **Greeting**  | Avatar greeting                   |