MXPROCESS

Asynchronous API: AI Video Publishing Pack

Upload one video with speech. The worker extracts audio with FFmpeg, transcribes it through OpenAI and produces captions, a thumbnail and reusable publication copy. Review generated copy before publication.

Endpoint: POST https://mxprocess.com/api/files/start-process
Authentication: Bearer JWT from POST /api/login_check
Cost: 20 tokens, charged after success.

Start the action

curl --location --request POST 'https://mxprocess.com/api/files/start-process' \
--header 'Authorization: Bearer YOUR_TOKEN' \
--form 'file=@"/path/to/product-launch.mp4"' \
--form 'actionName="VideoPublishingPack"' \
--form 'parameters=[["language",["English"]],["audience",["B2B buyers"]],["platform",["YouTube"]]]' \
--form 'async=true'

Accepted: MP4, MOV, MKV, WebM, AVI and M4V. Exactly one video, maximum 2 GB, with an audio track. Store id and poll /api/actions/get-action-infos/{id}.

{"id":123,"action":{"actionName":"VideoPublishingPack","slug":"ai-video-publishing-pack"},"accomplished":false,"trackingUrl":"https://mxprocess.com/api/actions/get-action-infos/123"}

Completed actionResults contains transcript TXT, captions SRT, publishing-pack Markdown and JSON, and a JPG thumbnail. Download through /api/actions/download-action-result/123/FILENAME.

Functional error example

{"error":"ERROR_VPP006 Video Publishing Pack requires a video with an audio track."}

For a client that can wait, use the synchronous API.