How to Use Stable Audio 3 Medium Base Audio to Audio in Codex
Use Stable Audio 3 Medium Base Audio to Audio to restyle recordings with a text description. Unlike Stable Audio 3 Medium Base Text to Audio, it starts from audio, with noise strength controlling how much of the source survives.
Prompts to try with Stable Audio 3 Medium Base Audio to Audio in Codex
Copy an example into Codex, or ask in your own words.
Use Stable Audio 3 Medium Base Audio to Audio to restyle the project's demo recording as warm electronic background music for 30 seconds. Start with noise strength 0.1 and guidance 7.
Use Stable Audio 3 Medium Base Audio to Audio on the project's existing score sketch to explore a darker orchestral treatment for 40 seconds. Compare noise strengths 0.1 and 0.5.
Compose a 30-second upbeat lo-fi track with Stable Audio 3 Medium Base Audio to Audio for a product video. Save the result in ./assets.
Make a calm piano loop with Stable Audio 3 Medium Base Audio to Audio for a meditation app.
Create a cinematic 20-second intro sting with Stable Audio 3 Medium Base Audio to Audio.
How to set up Stable Audio 3 Medium Base Audio to Audio in Codex
Pick the way that suits you. Each takes about two minutes.
Use Spiral's MCP server
- Add the hosted server
codex mcp add spiral --url https://api.spiral.ai/mcp
- Sign in
Codex opens the browser: sign in to Spiral and click Authorize. No API key. codex mcp list then shows Spiral.
codex mcp login spiral
Use the Spiral skill
- Send Codex 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 Stable Audio 3 Medium Base Audio to Audio
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/stable-audio-3/medium/base/audio-to-audio", "input": { "prompt": "bossa nova bassline", "audio_url": "https://storage.googleapis.com/falserverless/model_tests/diffrythm/rock_en.wav" } }'
Then paste one of the prompts above: Codex finds Stable Audio 3 Medium Base Audio to Audio on Spiral, shows the price and runs it.
Stable Audio 3 Medium Base Audio to Audio providers and prices
Spiral runs Stable Audio 3 Medium Base Audio to Audio through fal. Open a card for today's price, with every fee included.
Settings you can ask for
Stable Audio 3 Medium Base Audio to Audio works from your prompt and audio URL. Ask Codex for any of these in your prompt.
| Setting | Allowed values | Default | Description |
|---|---|---|---|
| Negative prompt | what to leave out, in words | Text description of qualities to avoid in the output. | |
| Duration | 1 to 380 | 30 | Duration of the generated audio in seconds. The medium model supports up to 380 seconds (~6m20s). |
| Number of inference steps | 1 to 100 | 50 | Number of sampling steps. Base (non-distilled) checkpoints typically need ~50 for good quality. |
| Guidance scale | 0 to 25 | 7 | Classifier-free guidance scale. Higher values follow the prompt more strictly; ~7.0 is a good starting point for base checkpoints. |
| Seed | a whole number | Random seed for reproducible outputs. Omit for a random seed. | |
| Enable prompt expansion | on or off | off | If True, the prompt will be expanded using an LLM for more detailed and higher quality results. |
| Output format | mp3, wav, flac, ogg, opus, m4a, or aac | mp3 | Container format for the generated audio output. |
| Init noise level | 0 to 1 | 0.9 | How much noise to add to the source audio. 1.0 fully replaces the audio with noise (pure generation); 0.1 keeps the source close to the input. 0.5 sits halfway between. |
Every API field of Stable Audio 3 Medium Base · Audio to Audio is on its tool page: open its card above.
What can you do with Stable Audio 3 Medium Base Audio to Audio in Codex?
Build instrumental audio drafts for game scenes, demos, and other project assets with Stable Audio 3 Medium Base. Use the project's creative brief to define the sound and review each result before adding it.
Podcast background music
Create or restyle a restrained instrumental bed with sparse piano, soft bass, and room for narration.
Product demo soundtracks
Develop a short instrumental cue for a walkthrough. Describe the instruments, pace, and energy the demonstration needs.
Game scene music drafts
Explore instrumental directions for a quiet village, tense corridor, or victory scene before selecting music for the game.
Short film score sketches
Draft a musical mood for a scene, using a written brief or an existing recording as the starting point.
Restyling a recorded idea
Start with a recording and explore a different musical character. Compare low and moderate noise strength to hear how much source detail survives.
More on Stable Audio 3 Medium Base Audio to Audio and Codex
Overview
Stable Audio 3 Medium Base Audio to Audio in other agents
Questions about Stable Audio 3 Medium Base Audio to Audio in Codex
Can Codex use Stable Audio 3 Medium Base Audio to Audio?
Yes. Add Spiral to Codex as an MCP server, and Codex can find and run Stable Audio 3 Medium Base Audio to Audio like any built-in tool.
How much does Stable Audio 3 Medium Base Audio to Audio cost in Codex?
$0.06799 per run at the default inputs. 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 Stable Audio 3 Medium Base Audio to Audio in Codex?
No. Spiral runs Stable Audio 3 Medium Base Audio 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 Stable Audio 3 Medium Base Audio to Audio in Codex?
About two minutes: add the hosted server, then sign in.
Does Codex wait while Stable Audio 3 Medium Base Audio to Audio runs?
Stable Audio 3 Medium Base Audio to Audio runs in the background. Codex starts the run, checks on it and gives you the audio when it's ready.
How long can Stable Audio 3 Medium Base audio be?
Up to 380 seconds, or 6 minutes and 20 seconds. Use a short clip for a short cue instead of requesting a full-length track.
Can Stable Audio 3 Medium Base sing lyrics?
It does not produce intelligible vocals. Vocal-like textures can still appear, so use it for instrumental work rather than songs that require understandable lyrics.
What does Base mean in Stable Audio 3 Medium Base?
It identifies the pretrained checkpoint intended for fine-tuning. Stability AI recommends the post-trained Stable Audio 3 Medium for direct generation; use roughly 50 sampling steps when generating with Base.
What audio quality does Stable Audio 3 Medium Base produce?
The underlying audio system uses stereo at 44.1 kHz. That describes the audio channels and sample rate, rather than guaranteeing that every generated track will sound finished.
Which language should I use for descriptions?
Use English as the documented starting point. The Medium Base model card lists English and does not establish reliable support for other prompt languages.
Start using Stable Audio 3 Medium Base Audio to Audio in Codex
One Spiral account for every tool your agent needs. No subscription; you pay per run.
Paste this into Codex:
set up https://spiral.ai/SKILL.md


