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

# Audio models

> The speech, music, and sound-effect models that power Orisu's audio nodes.

Audio models are selected on the audio-generating nodes. They cover three jobs: turning text into speech, composing music from a description, and adding sound effects to a video.

## Families

| Model                      | Provider   | Used by                                                | Job                 |
| -------------------------- | ---------- | ------------------------------------------------------ | ------------------- |
| ElevenLabs Multilingual v2 | ElevenLabs | [Generate Speech](/nodes/ai-audio/generate-speech)     | Text-to-speech      |
| ElevenLabs v3              | ElevenLabs | [Generate Speech](/nodes/ai-audio/generate-speech)     | Text-to-speech      |
| ElevenLabs Music v1        | ElevenLabs | [Generate Music](/nodes/ai-audio/generate-music)       | Music generation    |
| Mirelo 1.5                 | Mirelo     | [Add Sound Effects](/nodes/ai-audio/add-sound-effects) | Video sound effects |

## Settings

Each audio node exposes the settings relevant to its model. Availability depends on the model you pick.

### Speech

| Setting            | What it does                                                            |
| ------------------ | ----------------------------------------------------------------------- |
| `stability`        | Voice consistency, from 0 to 1. Higher values give a steadier delivery. |
| `similarity_boost` | How closely the output matches the chosen voice, from 0 to 1.           |
| `style`            | Style exaggeration, from 0 to 1.                                        |
| `speed`            | Playback speed, from 0.5 to 2.                                          |

### Music

| Setting              | What it does                                   |
| -------------------- | ---------------------------------------------- |
| `duration`           | Track length in seconds, from 3 to 600.        |
| `force_instrumental` | Generate an instrumental track with no vocals. |

See [Model settings](/models/settings) for the full reference.

## Related nodes

<Columns cols={2}>
  <Card title="Generate Speech" href="/nodes/ai-audio/generate-speech">
    Convert text to spoken audio with an ElevenLabs voice.
  </Card>

  <Card title="Generate Music" href="/nodes/ai-audio/generate-music">
    Compose a music track from a text description.
  </Card>

  <Card title="Add Sound Effects" href="/nodes/ai-audio/add-sound-effects">
    Generate and attach sound effects to a video.
  </Card>
</Columns>