Stable Audio 3 Audio Inpainting API

Stability AI Stable Audio 3 Audio-Inpainting

Stability AI Stable Audio 3 Audio-Inpainting replaces a selected region inside an existing audio clip using a natural-language prompt. It is designed for localized audio editing workflows where you want to rewrite only part of a clip while preserving the rest of the original audio.

Why Choose This?

Region-based audio replacement
Replace only the selected section of an audio clip instead of regenerating the whole file.
Prompt-guided inpainting
Use a text prompt to describe what the new audio inside the masked region should sound like.
Negative prompt support
Add negative_prompt to reduce unwanted sounds or qualities in the generated result.
Preserves surrounding audio
Audio outside the selected inpaint range remains unchanged.
Flexible output formats
Export results in mp3, wav, flac, ogg, opus, m4a, or aac.
Production-ready API
Suitable for sound repair, replacement, transition cleanup, ambient editing, and creative post-production workflows.

Parameters

Parameter	Required	Description
audio	Yes	Source audio to edit.
prompt	Yes	Text prompt describing the replacement audio for the masked region.
mask_start_seconds	No	Start time of the region to inpaint, in seconds. Default: `0`.
mask_end_seconds	No	End time of the region to inpaint, in seconds. Must be greater than `mask_start_seconds`. Default: `5`.
negative_prompt	No	Optional terms to avoid in the generated audio.
num_inference_steps	No	Number of inference steps. Range: `1–100`. Default: `8`.
guidance_scale	No	Prompt guidance strength. Range: `0–25`. Default: `1`.
output_format	No	Output audio format. Supported values: `mp3`, `wav`, `flac`, `ogg`, `opus`, `m4a`, `aac`. Default: `mp3`.

How to Use

Upload your source audio — provide the clip you want to edit.
Write your prompt — describe what the replacement audio inside the selected region should sound like.
Set the inpaint range — choose mask_start_seconds and mask_end_seconds to define the region to replace.
Add a negative prompt (optional) — list sounds or qualities you want to avoid.
Adjust generation controls (optional) — tune num_inference_steps and guidance_scale if needed.
Choose output format — select the format that best fits your workflow.
Submit — run the model and download the edited audio.

Example Prompt

Replace this section with a softer cinematic ambience, distant wind, subtle metallic resonance, and a smoother transition into the following audio.

Pricing

Just $0.0442 per request.

Billing Rules

Each inpainting request costs $0.0442
Pricing is fixed per request
mask_start_seconds, mask_end_seconds, negative_prompt, num_inference_steps, guidance_scale, and output_format do not affect pricing

Best Use Cases

Audio repair — Replace damaged or unwanted segments inside an existing clip.
Transition cleanup — Smooth over awkward cuts or rough edits in a timeline.
Sound redesign — Change one localized part of a scene without affecting the rest.
Creative post-production — Rewrite small sections of a soundtrack or ambience bed.
Effect replacement — Swap out a specific sound event while keeping surrounding context intact.

Pro Tips

Keep the masked region as tight as possible for more controlled replacement.
Use prompts that match the tone and texture of the surrounding audio for smoother blending.
Add a negative_prompt when you want to avoid music, vocals, distortion, or unwanted effects.
Increase num_inference_steps if you want potentially more refined results and can tolerate more runtime.
Choose wav or flac when you plan further editing after generation.

Notes

audio and prompt are required.
mask_start_seconds and mask_end_seconds define the exact region to replace.
mask_end_seconds must be greater than mask_start_seconds.
Audio outside the selected mask range is preserved.
Pricing is fixed at $0.0442 per request.

Related Models

Stability AI Stable Audio 3 Text-to-Audio — Generate audio directly from a text prompt.
Stability AI Stable Audio 3 Audio-Outpainting — Extend an existing audio clip before and/or after the source.
Other audio editing workflows — Useful when you need continuation, regeneration, or more specialized sound design controls.

Stable Audio 3 Audio Inpainting API — Quick start

Grab a WaveSpeedAI API key, then call POST https://api.wavespeed.ai/api/v3/stability-ai/stable-audio-3/audio-inpainting with your input as JSON. The endpoint returns a prediction id; poll the prediction endpoint until status flips to completed, then read the output URL from data.outputs[0]. Examples for Stable Audio 3 Audio Inpainting below.

HTTP example

# Submit the prediction
curl -X POST "https://api.wavespeed.ai/api/v3/stability-ai/stable-audio-3/audio-inpainting" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $WAVESPEED_API_KEY" \
  -d '{
    "audio": "https://example.com/your-audio.mp3",
    "prompt": "A cinematic shot of a city at sunset, soft golden light",
    "mask_start_seconds": 0,
    "mask_end_seconds": 5,
    "negative_prompt": "blurry, low quality, distorted",
    "num_inference_steps": 8,
    "guidance_scale": 1,
    "output_format": "mp3"
}'

# Response includes a prediction id. Poll for the result:
curl -X GET "https://api.wavespeed.ai/api/v3/predictions/{request_id}/result" \
  -H "Authorization: Bearer $WAVESPEED_API_KEY"

# When status is "completed", read the output from data.outputs[0].

Node.js example

// npm install wavespeed
const WaveSpeed = require('wavespeed');

const client = new WaveSpeed(); // reads WAVESPEED_API_KEY from env

const result = await client.run("stability-ai/stable-audio-3/audio-inpainting", {
        "audio": "https://example.com/your-audio.mp3",
        "prompt": "A cinematic shot of a city at sunset, soft golden light",
        "mask_start_seconds": 0,
        "mask_end_seconds": 5,
        "negative_prompt": "blurry, low quality, distorted",
        "num_inference_steps": 8,
        "guidance_scale": 1,
        "output_format": "mp3"
});

console.log(result.outputs[0]); // → URL of the generated output

Python example

# pip install wavespeed
import wavespeed

output = wavespeed.run(
    "stability-ai/stable-audio-3/audio-inpainting",
    {
    "audio": "https://example.com/your-audio.mp3",
    "prompt": "A cinematic shot of a city at sunset, soft golden light",
    "mask_start_seconds": 0,
    "mask_end_seconds": 5,
    "negative_prompt": "blurry, low quality, distorted",
    "num_inference_steps": 8,
    "guidance_scale": 1,
    "output_format": "mp3"
}
)

print(output["outputs"][0])  # → URL of the generated output

Stable Audio 3 Audio Inpainting API — Frequently asked questions

What is the Stable Audio 3 Audio Inpainting API?

Stable Audio 3 Audio Inpainting is a Stability AI model for AI inference, exposed as a REST API on WaveSpeedAI. Stable Audio 3 Audio Inpainting is a fast AI audio editing model that regenerates a selected region of an audio clip from a prompt while preserving the surrounding audio. Ready-to-use REST inference API for audio repair, sound effect editing, segment regeneration, music production, game audio, video sound design, and professional audio inpainting workflows with simple integration, no coldstarts, and affordable pricing. You can call it programmatically or try it from the playground above.

How do I call the Stable Audio 3 Audio Inpainting API?

POST your input parameters to the model's REST endpoint (shown in the API tab of this playground) with your WaveSpeedAI API key in the Authorization header. Submission returns a prediction ID; poll the prediction endpoint until status flips to "completed", then read the output URL from the result. The playground generates a ready-to-paste code sample in Python, JavaScript, or cURL for whatever inputs you've set. Full request/response shape is documented at https://wavespeed.ai/docs/docs-api/stability-ai/stability-ai-stable-audio-3-audio-inpainting.

How much does Stable Audio 3 Audio Inpainting cost per run?

Stable Audio 3 Audio Inpainting starts at $0.044 per run. That figure is the base price — the final charge scales with the parameters you set in the form (output size, length, count, references, or whatever knobs this model exposes), so a higher-quality or larger output costs more than a minimal one. The exact cost for your current input is shown live next to the Generate button before you submit, and the actual per-call charge is recorded on the prediction afterwards.

What inputs does Stable Audio 3 Audio Inpainting accept?

Key inputs: `prompt`, `audio`, `guidance_scale`, `num_inference_steps`, `negative_prompt`, `mask_end_seconds`. The full JSON schema (types, defaults, allowed values) is rendered above the Generate button and mirrored in the API reference at https://wavespeed.ai/docs/docs-api/stability-ai/stability-ai-stable-audio-3-audio-inpainting.

How do I get started with the Stable Audio 3 Audio Inpainting API?

Sign up for a free WaveSpeedAI account to claim starter credits, copy your API key from /accesskey, then call the endpoint shown in the API tab of the playground. The playground also auto-generates a code sample in Python, JavaScript, or cURL for the parameters you've set.

Can I use Stable Audio 3 Audio Inpainting outputs commercially?

Commercial usage rights depend on the model's license, set by its provider (Stability AI). The license summary appears on the model card above; see WaveSpeedAI's Terms of Service for platform-level conditions.

ExemplesTout voir

Modèles associés

README