[Integrations](https://spiral.ai/integrations) › [Codex](https://spiral.ai/integrations/codex) › [ACE-Step Prompt to Audio](https://spiral.ai/integrations/codex/ace-step-prompt-to-audio)

# How to Use ACE-Step Prompt to Audio in Codex

ACE-Step Prompt to Audio makes songs and instrumental tracks from a written music brief. It generates tags and lyrics from your description, making it useful for testing song ideas without writing lyrics first.

|  |  |
| --- | --- |
| Model | ACE-Step Prompt to Audio |
| Available through | fal |
| Category | Audio |
| Price | $0.00026 per second. Billed on `duration`, default 60. |
| Returns | audio |
| Runs | In the background |
| Setup | About 2 minutes, no API key |

## Prompts to try with ACE-Step Prompt to Audio in Codex

- Read the game's scene descriptions and use ACE-Step Prompt to Audio to generate a 60-second instrumental exploration theme with soft piano, acoustic guitar, and a calm mood. Save it with the music brief.
- Read the tutorial script in this project and use ACE-Step Prompt to Audio to make a 45-second instrumental backing track with sparse electric piano and light percussion. Keep the arrangement restrained for narration.
- Use ACE-Step Prompt to Audio to create three 30-second instrumental menu themes for this game: acoustic folk, quiet jazz, and ambient electronic. Save each take with its prompt and seed.
- Read the product demo storyboard and use ACE-Step Prompt to Audio to make a 60-second instrumental track with warm synths, bass, and steady drums. Match the storyboard's optimistic tone.
- Use ACE-Step Prompt to Audio to test the soundtrack brief in this project at 27 and 60 generation steps. Keep the prompt, seed, and 30-second duration fixed, and save both takes for listening review.

## How to set up ACE-Step Prompt to Audio in Codex

Pick the way that suits you. Each takes about two minutes.

### Use Spiral's MCP server

1. **Add the hosted server**

```bash
codex mcp add spiral --url https://api.spiral.ai/mcp
```

2. **Sign in**: Codex opens the browser: sign in to Spiral and click Authorize. No API key. `codex mcp list` then shows Spiral.

```bash
codex mcp login spiral
```

### Use the Spiral skill

1. **Send Codex this message**: Your agent reads Spiral's skill and sets itself up.

```
set up https://spiral.ai/SKILL.md
```

2. **Sign in or paste a key**: Approve Spiral in the browser, or give the agent a key from the console.

### Use the Spiral CLI

1. **Install and sign in**: Installs the Spiral CLI, signs you in and adds its skill to your coding agents.

```bash
npx -y spiral-cli@latest init
```

### Call the Spiral API

1. **Create an API key**: In the console, under API keys.

2. **Run ACE-Step Prompt to Audio**: One request; the result comes back by polling or a webhook.

```bash
curl https://api.spiral.ai/v1/runs \
  -H "Authorization: Bearer $SPIRAL_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: $(uuidgen)" \
  -d '{
    "provider": "fal",
    "model": "fal-ai/ace-step/prompt-to-audio",
    "input": {
      "prompt": "A lofi hiphop song with a chill vibe about a sunny day on the boardwalk."
    }
  }'
```

Then paste one of the prompts above: Codex finds ACE-Step Prompt to Audio on Spiral, shows the price and runs it.

Setup guide in the docs: https://spiral.ai/docs/quickstarts/codex

## ACE-Step Prompt to Audio tools and prices

- [ACE Step · Prompt to Audio](https://spiral.ai/tools/fal/fal-ai/ace-step/prompt-to-audio) (fal): $0.00026 per second. Billed on `duration`, default 60.

## Settings you can ask for

| Setting | Allowed values | Default | Description |
| --- | --- | --- | --- |
| instrumental |  | false | Whether to generate an instrumental version of the audio. |
| duration | 5 to 240 | 60 | The duration of the generated audio in seconds. |
| number_of_steps | 3 to 60 | 27 | Number of steps to generate the audio. |
| seed |  | — | Random seed for reproducibility. If not provided, a random seed will be used. |
| scheduler | euler, heun | euler | Scheduler to use for the generation process. |
| guidance_type | cfg, apg, cfg_star | apg | Type of CFG to use for the generation process. |

## What can you do with ACE-Step Prompt to Audio in Codex?

Create music drafts that match a game's scenes, a tutorial, or a product demo. A coding agent can read project briefs and organize ACE-Step outputs with the prompts and settings used for each take.

### Podcast music drafts

Generate a short instrumental opening with a clear genre and mood, then audition it against the spoken introduction.

### Tutorial background music

Request a restrained instrumental bed with sparse piano, soft percussion, and room for narration. Review the track alongside the voice recording.

### Game soundtrack prototypes

Turn descriptions of exploration, combat, or menu scenes into musical sketches. Compare different instrumentation before choosing a soundtrack direction.

### Song concept demos

Describe a song's subject, genre, and vocal character to hear an initial composition with generated lyrics. Use the result to judge the direction before further songwriting.

### Video music auditions

Create several instrumental takes for a travel montage or product demonstration. Keep the requested duration fixed while comparing musical approaches.

### Presentation soundtracks

Draft instrumental music for an event opening or presentation intermission. Specify the audience, mood, and instruments in the brief.

### Arrangement exploration

Try a musical idea with acoustic guitar, piano, or electronic instrumentation. Generate separate takes to hear which arrangement best serves the brief.

## Overview

- [fal](https://spiral.ai/providers/fal)
- [All Codex guides](https://spiral.ai/integrations/codex)

## ACE-Step Prompt to Audio in other agents

- [Amp](https://spiral.ai/integrations/amp/ace-step-prompt-to-audio)
- [ChatGPT](https://spiral.ai/integrations/chatgpt/ace-step-prompt-to-audio)
- [Claude](https://spiral.ai/integrations/claude-desktop/ace-step-prompt-to-audio)
- [Claude Code](https://spiral.ai/integrations/claude-code/ace-step-prompt-to-audio)
- [Cline](https://spiral.ai/integrations/cline/ace-step-prompt-to-audio)
- [Cursor](https://spiral.ai/integrations/cursor/ace-step-prompt-to-audio)
- [Devin](https://spiral.ai/integrations/devin/ace-step-prompt-to-audio)
- [dots](https://spiral.ai/integrations/dots/ace-step-prompt-to-audio)
- [Factory Droid](https://spiral.ai/integrations/factory-droid/ace-step-prompt-to-audio)
- [Gemini](https://spiral.ai/integrations/gemini/ace-step-prompt-to-audio)
- [Gemini CLI](https://spiral.ai/integrations/gemini-cli/ace-step-prompt-to-audio)
- [Google Antigravity](https://spiral.ai/integrations/antigravity/ace-step-prompt-to-audio)
- [goose](https://spiral.ai/integrations/goose/ace-step-prompt-to-audio)
- [Grok](https://spiral.ai/integrations/grok/ace-step-prompt-to-audio)
- [Grok Build](https://spiral.ai/integrations/grok-build/ace-step-prompt-to-audio)
- [Hermes](https://spiral.ai/integrations/hermes/ace-step-prompt-to-audio)
- [Instinct](https://spiral.ai/integrations/instinct/ace-step-prompt-to-audio)
- [Manus](https://spiral.ai/integrations/manus/ace-step-prompt-to-audio)
- [Muse](https://spiral.ai/integrations/muse/ace-step-prompt-to-audio)
- [Muse Code](https://spiral.ai/integrations/muse-code/ace-step-prompt-to-audio)
- [n8n](https://spiral.ai/integrations/n8n/ace-step-prompt-to-audio)
- [Notion](https://spiral.ai/integrations/notion/ace-step-prompt-to-audio)
- [OpenClaw](https://spiral.ai/integrations/openclaw/ace-step-prompt-to-audio)
- [opencode](https://spiral.ai/integrations/opencode/ace-step-prompt-to-audio)
- [Perplexity](https://spiral.ai/integrations/perplexity/ace-step-prompt-to-audio)
- [Raycast](https://spiral.ai/integrations/raycast/ace-step-prompt-to-audio)
- [VS Code](https://spiral.ai/integrations/vscode/ace-step-prompt-to-audio)
- [Windsurf](https://spiral.ai/integrations/windsurf/ace-step-prompt-to-audio)
- [Zapier](https://spiral.ai/integrations/zapier/ace-step-prompt-to-audio)

## More Codex guides

- [ACE-Step Lyrics to Audio](https://spiral.ai/integrations/codex/ace-step-lyrics-to-audio)
- [Async Pro 1.0](https://spiral.ai/integrations/codex/async-pro-1-0)
- [ACE-Step Audio to Audio](https://spiral.ai/integrations/codex/ace-step-audio-to-audio)
- [Audio Understanding](https://spiral.ai/integrations/codex/audio-understanding)
- [ACE-Step Audio Outpaint](https://spiral.ai/integrations/codex/ace-step-audio-outpaint)
- [Canary Speech-to-Text](https://spiral.ai/integrations/codex/canary-speech-to-text)

## Questions about ACE-Step Prompt to Audio in Codex

**Can Codex use ACE-Step Prompt to Audio?**

Yes. Add Spiral to Codex as an MCP server, and Codex can find and run ACE-Step Prompt to Audio like any built-in tool.

**How much does ACE-Step Prompt to Audio cost in Codex?**

$0.00026 per second. Billed on `duration`, default 60. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Codex plan is separate.

**Do I need a fal account to use ACE-Step Prompt to Audio in Codex?**

No. Spiral runs ACE-Step Prompt to Audio on its own fal account. You sign in to Spiral once and never paste a key.

**How long does it take to set up ACE-Step Prompt to Audio in Codex?**

About two minutes: add the hosted server, then sign in.

**Does Codex wait while ACE-Step Prompt to Audio runs?**

ACE-Step Prompt to Audio runs in the background. Codex starts the run, checks on it and gives you the audio when it's ready.

**Can ACE-Step make songs with vocals and instrumental music?**

Yes. ACE-Step Prompt to Audio turns a music brief into tags and lyrics for a song, or you can request an instrumental track.

**How long can ACE-Step tracks be?**

ACE-Step supports tracks up to four minutes long. Compare takes at the same duration when testing a prompt, because changing the length can also change the musical result.

**What should I put in an ACE-Step prompt?**

Describe the genre, main instruments, mood, and intended use. Short tags and complete sentences both work; for example, ask for a restrained jazz instrumental beneath a spoken tutorial.

**Which languages does ACE-Step support?**

ACE-Step supports 19 languages. The developers name English, Chinese, Russian, Spanish, Japanese, German, French, Portuguese, Italian, and Korean as the ten strongest.

**Where does ACE-Step need the most listening review?**

Listen closely to vocal expression and unusual instruments. The developers flag limited vocal nuance and imperfect rendering of rare instruments, so treat these outputs as drafts to review.

