# audio

> Reference for the audio tool of AI Audio: create, runs, list, get, delete, usage and providers, with every action.

Source: https://aiaudio.geenius.io/docs/reference/audio/

Sound from words in a workspace. A run you queue through the door is made as soon as somebody in that workspace has AI Audio open, because the Preview provider runs in a browser. Server providers run on their own once their keys are set.

Call it as `geenius_geeniusaudio(tool: "audio", action: "<action>", args: {...})` through Geenius OS.

## Actions

| Action | What it does |
|---|---|
| `create` | Queues a run of 1 to 4 takes of a sound and answers the run's id, the credits it costs and the app address. Speech: what to say. Dialogue: a script, one line per turn as "Ava: ..." with exactly two speakers. Music or effect: a description. Language, voice, speed (0.5 to 2), emotion, mood, seconds and loop depend on the kind. Premium and Max only. |
| `runs` | The latest runs with their state (queued, running, done or failed) and takes. |
| `list` | Sounds; a query searches words in the prompt. |
| `get` | One sound with its file address, length and waveform. |
| `delete` | A run and its takes for good; confirm with the run id. |
| `usage` | Credits used this month against the plan, and whether the plan can make sound. |
| `providers` | The models, what each can do and whether it is ready here. |

> [!TIP]
> Ask for the exact arguments of any action with `action: "help", of: "create"`. The server answers from its own definitions, so that answer is always current.

> [!NOTE]
> Making sound is the AI feature: Premium and Max only.