> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.empiriolabs.ai/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.empiriolabs.ai/_mcp/server.

# StepAudio 3 Gen

> Generates dialogue, sound effects, ambience, and background music together in one clip from a written scene.

![StepAudio 3 Gen](https://media.empiriolabs.ai/model-logos/stepaudio-3-gen.png)

[StepFun](/providers/stepfun) · Audio Generation

`POST /v1/audio/generations`

Generates dialogue, sound effects, ambience, and background music together in one clip from a written scene.

## At a glance

| Field               | Value                                                          |
| ------------------- | -------------------------------------------------------------- |
| Model id            | `stepaudio-3-gen`                                              |
| Model release date  | 2026-09-09                                                     |
| Input modalities    | Text                                                           |
| Output modalities   | Audio                                                          |
| Context window      | -                                                              |
| Weight precision    | -                                                              |
| Region              | International                                                  |
| Features            | audio_generation, text_to_speech, sound_effects, voice_control |
| Native inference    | No                                                             |
| New                 | Yes                                                            |
| Supported endpoints | `POST /v1/audio/generations`                                   |
| Alternate model ids | `stepaudio-3-gen-preview`, `stepfun/stepaudio-3-gen`           |

## Pricing

| Charge     | Spec                  | Rate   |
| ---------- | --------------------- | ------ |
| Generation | per 10,000 characters | \$0.36 |

## Example request

```bash
curl https://api.empiriolabs.ai/v1/audio/generations \
  -H 'Authorization: Bearer $EMPIRIOLABS_API_KEY' \
  -H 'Content-Type: application/json' \
  -d '{"model": "stepaudio-3-gen", "instruction": "A quiet kitchen in the morning, with rain on the window.", "prompt": "Good morning. The coffee is ready."}'
```

## Parameters

| Parameter            | Type   | Required | Default      | Description                                                                                                                                                                                |
| -------------------- | ------ | -------- | ------------ | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ |
| `prompt`             | string | no       | -            | Text to speak, read as written. Wrap a sound effect or music cue in square brackets and a delivery direction in parentheses, and describe the setting and background music in instruction. |
| `instruction`        | string | no       | -            | Global guidance for the setting, background music, and emotional tone. Maximum 500 characters. · Max: 500                                                                                  |
| `roles`              | string | no       | -            | Optional speakers, as objects with a name and a voice description. All names and descriptions together are capped at 500 characters.                                                       |
| `scripts`            | string | no       | -            | Optional ordered lines, as objects with a speaker and text. Capped at 1,000 characters in total. Use this instead of prompt for multi-speaker scenes.                                      |
| `response_format`    | enum   | no       | `"mp3"`      | Output audio format. · Allowed: `mp3`, `wav`, `flac`, `opus`, `pcm`                                                                                                                        |
| `sample_rate`        | enum   | no       | `24000`      | Output sample rate in Hz. · Allowed: `8000`, `16000`, `22050`, `24000`, `48000`                                                                                                            |
| `speed`              | number | no       | `1`          | Speech speed. · Range: 0.5 – 2                                                                                                                                                             |
| `volume`             | number | no       | `1`          | Output volume. · Range: 0.1 – 2                                                                                                                                                            |
| `text_normalization` | enum   | no       | `"standard"` | How numbers, dates, and symbols are read aloud. · Allowed: `standard`, `enhanced`                                                                                                          |

## Notes

One finished clip with spoken lines, sound effects, ambience, and background music together. Text in prompt and scripts is spoken as written: describe the setting and background music in instruction, wrap a sound effect or music cue in square brackets, and put a delivery direction in parentheses. Send a prompt, or use roles and scripts for multi-speaker control. Roles and instruction are capped at 500 characters each and scripts at 1,000. Output formats are mp3, wav, flac, opus, and pcm. Reference voices and voice cloning are not available on this model. This model is in preview and its capabilities may change.

---

*Machine-readable schema:* `GET https://api.empiriolabs.ai/v1/models/stepaudio-3-gen`.