GET

/v1/schemas/stability/audio/v2beta%2Faudio%2Fstable-audio%2Faudio-to-audio

kind inputactivity audiohash 46db054d…extracted Oct 9, 2026version currentsource upstream spec ↗

Request body

9 properties · 2 required · 2 $defs
propertytypedescriptionconstraints
prompt*stringWhat you wish the output audio to be. A strong, descriptive prompt that clearly defines instruments, moods, styles, and genre will lead to better results.length ≤ 10000
audio*stringThe audio to be used as the starting point for the generation.format: binary
modelenumThe model to use for generation.stable-audio-3 · default: "stable-audio-3"
durationnumberControls the duration in seconds of the generated audio.1 ≤ n ≤ 380 · default: 190
seednumberA specific value that is used to guide the 'randomness' of the generation. (Omit this parameter or pass `0` to use a random seed.)0 ≤ n ≤ 4294967294 · default: 0
stepsintegerControls the number of sampling steps.4 ≤ n ≤ 8 · default: 8
cfg_scalenumberHow strictly the diffusion process adheres to the prompt text (higher values make your audio closer to your prompt). Defaults to 1 if not specified.1 ≤ n ≤ 25 · default: 1
output_formatenumDictates the `content-type` of the generated audio.mp3 | wav · default: "mp3"
strengthnumberSometimes referred to as _denoising_, this parameter controls how much influence the `audio` parameter has on the generated audio. A value of 0 would yield audio that is identical to the input. A value of 1 would be as if you passed in no audio at all.0 ≤ n ≤ 1 · default: 1

Validate a payload

before you spend tokens
shell
$ curl -X POST https://modelschemas.com/v1/validate \
    -d '{"provider":"stability","endpointId":"v2beta/audio/stable-audio/audio-to-audio","payload":{…}}'
→ {"valid": false, "errors": [{"path": "#", "keyword": "required", …}]}