GET

/v1/schemas/fal/video/veed/dubbing

Accept:

kind inputactivity videohash b32f7d0b…extracted Nov 13, 2026version currentsource upstream spec ↗

field	recorded source
schema	Source linked above. upstream-spec Source trace `0ea98ab185777267c3e4a8271b2e82b958b186f6254c0bd45e9a46045492d3d8` { "fetchedAt": 1794546001, "extractorVersion": "2", "verifiedAt": null }

Request body

4 properties · 2 required · 0 $defs

property	type	description	constraints
target_language*	enum	Language to dub into. This model has no regional dialects: a regional tag such as 'fr-FR' is accepted and dubbed in the base language ('fr').	af \| ar \| as \| ast \| az \| be · +80 more
video_url*	string	Video to dub, 8 seconds to 10 minutes long and with speech in its audio track. The speech is translated into the target language, and the speakers' lips are synced to the dubbed voices.	length ≥ 1 · length ≤ 2083 · format: uri
keyterms	string[]	Names and terms to get right in the dub, such as brand, product or people's names: they help the speech recognition catch these words and the translation keep them as written. Up to 1,000 terms, each at most 50 characters and 5 words, without the characters < …	items ≤ 1000 · default: []
source_language	enum	Language spoken in the video. 'auto' (the default) detects it; set it explicitly if detection picks the wrong language.	auto \| af \| ar \| as \| ast \| az · +81 more · default: "auto"

Validate a payload

before you spend tokens

shell

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