Lip-sync face identification
Lip sync (`kling-lip-sync`) step one: identify faces in the media and get a session ID and face IDs.
/v1/kling/facesSynchronous endpoint, billed per call. The result contains a session ID and face IDs, valid for 24 hours. In step two, put them in extra.session_id and extra.face_choose (exactly one item, with face_id and sound_file) of Submit video task; see Lip sync for the flow.
The request body accepts only a videos array; each item has either url or file_id, not both, following the same rules as Media locators.
Headers
Bearer sk-…: an API key created in the console under API keys; its group must include the requested model
Body parametersJSON
Media to identify faces in; at least one item
Items≥ 1
Publicly accessible absolute HTTP(S) address; cannot be an internal, loopback or link-local address, or a URL with credentials
File ID of an existing asset
Returns
200Identification result JSON with a session ID and face IDs (valid for 24 hours)
application/json Face identification result
Errors
request body does not match the JSON contract: …: invalid JSON, a field outside the table, or a second JSON value after the request body
videos is required and must contain at least one item
videos[<index>] must provide exactly one of url or file_id
videos[<index>].url must be a publicly accessible absolute http(s) URLmissing_api_key / invalid_api_key / api_key_expired)