// Transcribe video files directly into timed transcripts and subtitle-ready artifacts using hosted Whisper video-to-text. Use this when the input is a video and the goal is speech extraction, caption generation, or edit-prep timing.
[HINT] Download the complete skill directory including SKILL.md and all related files
| name | video-transcription |
| description | Transcribe video files directly into timed transcripts and subtitle-ready artifacts using hosted Whisper video-to-text. Use this when the input is a video and the goal is speech extraction, caption generation, or edit-prep timing. |
| metadata | {"postplus":{"familyId":"media-production","familyName":"Media and Creative Production"}} |
Follow shared public skill rules in:
postplus-shared public skill rulesUse this skill when the input is video and the main problem is:
This skill is not a substitute for semantic video analysis.
First-version hosted transcription endpoint:
transcription-whisper-with-videoUse enableTimestamps=true by default when the result will drive subtitles or edit decisions.
Hosted video transcription is asynchronous. The script polls the prediction result URL
until status is completed or failed. Default poll window: 150 attempts × 2 s = 5 minutes.
Real speech video typically completes within 30–60 s. Silent or very short videos may complete in under 10 s. If a job exceeds 5 minutes, the hosted provider is likely overloaded — retry rather than increasing the timeout further.
Before submission, the script logs a polling preflight line from
durationSeconds. Video at or above 300 seconds is marked as possibly exceeding
the current polling window. The script does not silently increase the timeout or
switch tools; the current contract remains a 5-minute poll followed by a timeout
failure if the provider has not completed.
Before submission:
durationSeconds from the source videodurationSeconds in the request JSONIf the request needs guaranteed processing beyond the current polling window, stop before submission and say that the released transcription contract does not guarantee that path. Do not silently increase the timeout or switch to an unsupported transcription tool.
Persist:
request.jsonresponse.jsonmanifest.jsonoutputs/scripts/transcribe_video.mjsPolling support is shared with:
skills/40-creative/audio-transcription/scripts/poll_transcription.mjsreferences/tool-contracts.mdreferences/normalized-transcript-schema.md<work-folder>/.postplus/video-transcription/.postplus/