GET

/v1/schemas/fal/audio/mirelo-ai/sfx2.0/video-to-audio?kind=output

Accept:

kind outputactivity audiohash b7b08636…extracted Oct 22, 2026version currentsource upstream spec ↗

field	recorded source
schema	Source linked above. upstream-spec Source trace `df9d8d659ad29aaabe474218810b66902bf73f93bf691a79609f9189121a4652` { "fetchedAt": 1792645225, "extractorVersion": "1", "verifiedAt": null }

Response body

3 properties · 2 required · 2 $defs

property	type	description	constraints
model*	string	The SFX model release that made the output
audio*	Audio[]	Each generated soundtrack as an audio file, without the video
speech	ReturnedSpeech	What a return_speech request found in the input's speech, and the file made of it.	nullable

Types

2 $defs

Audioobject

content_type	string	The mime type of the file.	nullable
file_size	integer	The size of the file in bytes.	nullable
file_name	string	The name of the file. It will be auto-generated if not provided.	nullable
url*	string	The URL where the file can be downloaded from.

ReturnedSpeechobject

What a return_speech request found in the input's speech, and the file made of it.

windows_failed*	integer	Windows whose separation failed, or whose stem could not be read
outcome*	enum	speech when audio holds the file, no_speech when no window of the input has speech (or it has no audio), failed when separating a window or building the file failed. The soundtracks come back either way.	speech \| no_speech \| failed
windows*	integer	Windows the input was separated in
audio	Audio	The separated speech as one audio file; null unless the outcome is speech	nullable
windows_with_speech*	integer	Windows that held speech

Validate a payload

before you spend tokens

shell

$ curl -X POST https://modelschemas.com/v1/validate \
    -d '{"provider":"fal","endpointId":"mirelo-ai/sfx2.0/video-to-audio","payload":{…}}'
→ {"valid": false, "errors": [{"path": "#", "keyword": "required", …}]}