How to Use ACE-Step Audio Inpaint in ChatGPT
ACE-Step Audio Inpaint regenerates a selected passage in existing music using style descriptions or lyrics. Use it to try a different chorus, vocal phrase, or instrumental passage while retaining the surrounding audio.
Prompts to try with ACE-Step Audio Inpaint in ChatGPT
Copy an example into ChatGPT, or ask in your own words.
Use ACE-Step Audio Inpaint on the supplied song from 20–28 seconds. Try acoustic folk with fingerpicked guitar, gentle percussion, and soft vocals; use the existing lyrics.
Use ACE-Step Audio Inpaint on the supplied instrumental from 10–18 seconds. Request mellow jazz piano, double bass, brushed drums, and no sung lyrics.
Use ACE-Step Audio Inpaint on the supplied song from 30–38 seconds. Replace the chorus words with '[chorus] We found our way back home / Beneath the morning sky' and request warm pop vocals.
Use ACE-Step Audio Inpaint on the supplied electronic track from 16–24 seconds. Try restrained synths, soft bass, light percussion, and no sung lyrics; make three auditions for me to compare.
Use ACE-Step Audio Inpaint on the supplied vocal demo from 12–20 seconds. Keep the existing words and request a gentler vocal delivery with acoustic guitar.
How to set up ACE-Step Audio Inpaint in ChatGPT
Pick the way that suits you. Each takes about two minutes.
Use the ChatGPT connector
- Add a custom MCP server
In ChatGPT open Plugins, then Add › Add custom MCP server. Name it Spiral, paste this URL and keep Authentication on OAuth. Tick I understand and want to continue, then click Create.
https://api.spiral.ai/mcp
- Click Connect
ChatGPT asks to connect Spiral. Connect takes you to Spiral in your browser.
- Sign in to Spiral and approve
Sign in, pick the organization ChatGPT should use and click Authorize. Spiral gives the connection its own API key; ChatGPT never sees it, and you can remove the connection under Settings › Connected apps.
Use the Spiral skill
- Send ChatGPT this message
Your agent reads Spiral's skill and sets itself up.
set up https://spiral.ai/SKILL.md
- Sign in or paste a key
Approve Spiral in the browser, or give the agent a key from the console.
Use the Spiral CLI
- Install and sign in
Installs the Spiral CLI, signs you in and adds its skill to your coding agents.
npx -y spiral-cli@latest init
Call the Spiral API
- Create an API key
In the console, under API keys.
- Run ACE-Step Audio Inpaint
One request; the result comes back by polling or a webhook.
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/audio-inpaint", "input": { "audio_url": "https://storage.googleapis.com/falserverless/example_inputs/ace-step-audio-to-audio.wav", "tags": "lofi, hiphop, drum and bass, trap, chill" } }'
Then paste one of the prompts above: ChatGPT finds ACE-Step Audio Inpaint on Spiral, shows the price and runs it.
ACE-Step Audio Inpaint providers and prices
Spiral runs ACE-Step Audio Inpaint through fal. Open a card for today's price, with every fee included.
Settings you can ask for
ACE-Step Audio Inpaint works from your audio URL. Ask ChatGPT for any of these in your prompt.
| Setting | Allowed values | Default | Description |
|---|---|---|---|
| Start time relative to | start or end | start | Whether the start time is relative to the start or end of the audio. |
| Start time | 0 to 240 | 0 | start time in seconds for the inpainting process. |
| End time relative to | start or end | start | Whether the end time is relative to the start or end of the audio. |
| End time | 0 to 240 | 30 | end time in seconds for the inpainting process. |
| Variance | 0 to 1 | 0.5 | Variance for the inpainting process. Higher values can lead to more diverse results. |
| Number of steps | 3 to 60 | 27 | Number of steps to generate the audio. |
| Seed | a whole number | Random seed for reproducibility. If not provided, a random seed will be used. | |
| Scheduler | euler or heun | euler | Scheduler to use for the generation process. |
| Guidance type | cfg, apg, or cfg_star | apg | Type of CFG to use for the generation process. |
| Granularity scale | -100 to 100 | 10 | Granularity scale for the generation process. Higher values can reduce artifacts. |
| Guidance interval | 0 to 1 | 0.5 | Guidance interval for the generation. 0.5 means only apply guidance in the middle steps (0.25 * infer_steps to 0.75 * infer_steps) |
| Guidance interval decay | 0 to 1 | 0 | Guidance interval decay for the generation. Guidance scale will decay from guidance_scale to min_guidance_scale in the interval. 0.0 means no decay. |
| Guidance scale | 0 to 200 | 15 | Guidance scale for the generation. |
| Minimum guidance scale | 0 to 200 | 3 | Minimum guidance scale for the generation after the decay. |
| Tag guidance scale | 0 to 10 | 5 | Tag guidance scale for the generation. |
| Lyric guidance scale | 0 to 10 | 1.5 | Lyric guidance scale for the generation. |
Every API field of ACE Step Edit · Audio Inpaint is on its tool page: open its card above.
What can you do with ACE-Step Audio Inpaint in ChatGPT?
Explore a new musical direction for a passage in supplied audio with ACE-Step Audio Inpaint. A chat app can turn your description into a focused edit brief and help you compare auditions.
Audition a chorus rewrite
Try revised lyrics in the chorus of an existing song demo and compare the phrasing before committing to a rewrite.
Rework an instrumental break
Regenerate a selected instrumental passage with a different musical direction, such as piano instead of a synth-led arrangement.
Test a quieter podcast music bed
Try a gentler arrangement in the passage that sits beneath narration in an existing podcast theme.
Explore a game music transition
Create alternative versions of a short passage in an existing game cue to audition a change in mood or intensity.
Revise a sung campaign line
Test new wording in a selected phrase of an existing musical ad demo, then review intelligibility and timing.
Compare vocal directions
Regenerate a short sung passage with different vocal descriptions to explore how the performance changes.
More on ACE-Step Audio Inpaint and ChatGPT
Overview
ACE-Step Audio Inpaint in other agents
Questions about ACE-Step Audio Inpaint in ChatGPT
Can ChatGPT use ACE-Step Audio Inpaint?
Yes. Add Spiral to ChatGPT as an MCP server, and ChatGPT can find and run ACE-Step Audio Inpaint like any built-in tool.
How much does ACE-Step Audio Inpaint cost in ChatGPT?
Up to $0.0156 per run at the default inputs for 60 seconds of input. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your ChatGPT plan is separate.
Do I need a fal account to use ACE-Step Audio Inpaint in ChatGPT?
No. Spiral runs ACE-Step Audio Inpaint 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 Audio Inpaint in ChatGPT?
About two minutes: add a custom MCP server, then click Connect, then sign in to Spiral and approve.
Does ChatGPT wait while ACE-Step Audio Inpaint runs?
ACE-Step Audio Inpaint runs in the background. ChatGPT starts the run, checks on it and gives you the audio when it's ready.
What can ACE-Step Audio Inpaint change?
It can regenerate a selected passage using new lyrics or a style description. Useful targets include a chorus, a vocal phrase, or an instrumental passage.
Does ACE-Step Audio Inpaint preserve the surrounding song?
Repainting is designed to preserve audio outside the selected region. Listen across both edit boundaries before accepting a result.
What should I put in the style description?
Specify the genre, main instruments, approximate tempo, mood, and desired vocal character. For example: acoustic folk, fingerpicked guitar, 90 BPM, gentle, soft vocals.
Can ACE-Step Audio Inpaint make an instrumental passage?
Yes. Use [inst] or [instrumental] to request a passage without sung lyrics, and describe the instruments and mood separately.
How consistent are repeated edits?
The maker reports strong sensitivity to random seeds and input duration. Compare several attempts on the same passage before changing the edit boundaries.
Start using ACE-Step Audio Inpaint in ChatGPT
One Spiral account for every tool your agent needs. No subscription; you pay per run.
Paste this into ChatGPT:
set up https://spiral.ai/SKILL.md


