Translate and caption
Translate a transcript into many languages with cue-accurate captions, and optionally dubbed audio. Priced per language-minute. One request can be twenty languages of a ten-minute cut — two independent multipliers, so both move the price.
Authorizations
A workspace API key.
Headers
Stable retry key for this logical paid call. Reuse it after a timeout to receive the original result without duplicate work or spend.
1 - 255Send 1 to get the price, balance and budget verdict without running anything.
"1"The approved price; a call that would cost more is refused and nothing is charged.
^\d+$Body
- Option 1
- Option 2
Send exactly one source: either transcript or assetId, not both and not neither. Priced per language-minute of the transcript that results.
The transcript to translate. Send exactly one of transcript or assetId.
One to 20 BCP 47 language tags to produce, e.g. ["es", "sw"].
1 - 20 elements1 - 20The existing project to file this work under. Create it with POST /v1/projects, then reuse its id across related calls. A project this workspace does not own answers not_found; the work is never silently filed somewhere else.
1 - 200Translate a transcript this workspace has already had transcribed instead of sending one. Send exactly one of transcript or assetId. Asset ids come from GET /v1/assets (an asset is usable once transcriptReady is true). An asset in another workspace is reported as not found.
1 - 200Source language. Required with assetId when no language was recorded for that asset — there is no default, because guessing would mis-translate a non-English asset silently.
1 - 20Also synthesize dubbed audio. Defaults to false; dubbing costs more than captions.
Caption file format. Omit to use the service default.
srt, vtt Provider-specific narration voice id used only when dub is true. Omit to use the deployment default; the public API has no universal cross-provider voice list.
120Premium dubbing provider. English dubs default to Speechify Simba 3.2. Non-English dubs require an explicit compatible provider; providers are never auto-switched.
speechify, elevenlabs, deepgram, gemini Provider-specific approved model. Set provider whenever model is supplied.
"simba-3.2"Translation context such as product names, preferred terminology, and phrases that must not be translated.
4000Response
Success.
- Option 1
- Option 2
What this call actually cost, which is not always what was quoted.
300
Per-language captions with the cue timing preserved, plus dubbed audio if asked.
What was billed: languages multiplied by transcript minutes.
The existing project this work was filed under. It matches the projectId supplied by the caller after workspace ownership was verified.
"prj_9x2a4c"
Your balance after the charge. Absent when billing is disabled for the workspace.
3900
Present only when assetId was sent: the indexed transcript this call reused, how many cues, and when it was captured.