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

# Text To Speech

The **Text-to-Speech** dashboard submits paid audio generations through Audixa's Base and Advanced models.

## The Generation Workflow

1. **Select Your Model:** Choose **Base** for its speed control or **Advanced** for `cfg_weight` and `exaggeration` controls.
2. **Choose Your Voice:** Pick from our extensive library of preset voices or use your own **Clone Voices**.
3. **Enter Your Text:** Type or paste your script.
4. **Settings (Optional):** Adjust speed, language, or advanced controls.
5. **Generate:** Create your audio.

## Model Selection

You can switch between two models depending on your needs:

* **Base Model:** Uses Base-compatible voices and exposes speed control.
* **Advanced Model:** Uses Advanced-compatible voices and exposes guidance and expressiveness controls. Required for **Clone Voices**.

## Quick Presets

To help you get started, we provide several one-click presets for common use cases:

* **Casual Conversation**
* **Professional Narration**
* **Short Story**
* **Announcement**

***

## Settings

### Voice Selection

Click the voice selector to open the Voice Modal. You can filter by:

* **Accent**
* **Gender** (Male, Female, Neutral)
* **Free Voices**

The “Free Voices” label is a voice-access classification used by plan rules. It
does not make a generation free.

### Base Model Settings

When the Base Model is selected, you can adjust:

* **Language:** Select from available languages (e.g., American English).
* **Speed:** Adjust playback speed from **0.5x** to **2.0x**.

### Advanced Model Settings

These settings provide granular control over the vocal performance and are only available with the Advanced Model.

| Setting | Description |
| :- | :- |
| **CFG Weight** | Controls the guidance strength (1.0 - 5.0). Higher values force the model to follow the text and style more strictly. |
| **Exaggeration** | Controls the expressiveness (0.0 - 1.0). Higher values result in more dramatic emotional fluctuation. |

### Output Format

Choose your preferred file format:

* **MP3:** Compressed, smaller file size (standard).
* **WAV:** Uncompressed output.

***

## Generating & Playing Audio

Once generated, the audio player appears at the bottom.

* **Waveform:** Visual representation of the audio.
* **Play/Pause/Download:** Standard controls.
* **History:** Your recent generations appear in the sidebar list.

## Automate with the API

You can automate this entire workflow using our API.

* [**API Introduction**](/api-reference/introduction)
* [**Generate Speech Route**](/api-reference/tts)


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.