GET

/v1/schemas/fal/video/fal-ai/controlfoley

Accept:

kind inputactivity videohash 2d980912…extracted Jul 6, 2026version currentsource upstream spec ↗

field	recorded source
schema	Source linked above. Source trace `1e934e5073b43457f8e49bd79d3f7d6d8f0665b9d28de1f96c438fcf8afeaa39` { "fetchedAt": 1783330412, "extractorVersion": "1", "verifiedAt": null }

Request body

9 properties · 1 required · 0 $defs

property	type	description	constraints
video_url*	string	URL of the video to generate synchronized audio for.
negative_prompt	string	Negative text prompt — describe audio characteristics to avoid.	default: ""
guidance_scale	number	Classifier-free guidance strength.	0 ≤ n ≤ 20 · default: 4.5
prompt	string	Optional text prompt describing the desired audio. When combined with the video it provides text-controlled video-to-audio (TC-V2A) generation; leave empty for pure video-to-audio (V2A).	default: ""
duration	number	Target audio duration in seconds. Truncated to source video length when shorter.	1 ≤ n ≤ 30 · default: 8
seed	integer	Random seed for reproducibility. If null, a random seed is used.	nullable
reference_audio_url	string	Optional URL to a 2-4 second reference audio clip whose timbre should be transferred to the generated audio (AC-V2A mode). Audio shorter than 2s is zero-padded; longer is truncated.	nullable
num_inference_steps	integer	Number of flow-matching ODE steps.	4 ≤ n ≤ 100 · default: 25
mask_away_clip	boolean	When True, disables the CLIP visual stream (text-driven mode).	default: false

Validate a payload

before you spend tokens

shell

$ curl -X POST https://modelschemas.com/v1/validate \
    -d '{"provider":"fal","endpointId":"fal-ai/controlfoley","payload":{…}}'
→ {"valid": false, "errors": [{"path": "#", "keyword": "required", …}]}