Turn long screen recordings into course modules

Course businesses sitting on hours of raw screen-recorded teaching. A three-hour screen recording is the most valuable asset a course business owns and the least usable. Nobody re-watches it to find the one part worth turning into a lesson, so it stays on a drive.

The flow

1. TriggerA new recording lands in the Drive folder your team records into.
2. SubmitYour scenario POSTs the file URL to /api/v1/transcribe/url with word_timestamps and an Idempotency-Key set to the Drive file id. The scenario ends here — it does not wait; the job_id goes to a store the poller reads.
3. PollStore the job_id and poll GET /api/v1/jobs/{id} from a scheduled run. Self-serve keys are Pro tier and collect by polling; signed callbacks are Enterprise.
4. SplitWord timestamps give you the boundaries. Section breaks fall where the narration changes topic, and each becomes a module with a real start and end time.
5. WriteThe transcript and outline write into your course tool; the timestamps drive the chapter markers.

# submit — returns immediately, does not block
curl -X POST https://whipscribe.com/api/v1/transcribe/url \
  -H "X-API-Key: $KEY" \
  -H "Idempotency-Key: screen-recordings-to-course-modules-$SOURCE_ID" \
  -H "Content-Type: application/json" \
  -d '{"url":"https://example.com/recording.mp4",
       "diarize":true,
       "word_timestamps":true}'

→ 202 {"job_id":"a1b2…","status":"queued"}

# then poll — a self-serve (Pro) key collects by polling
curl "https://whipscribe.com/api/v1/jobs/$JOB" -H "X-API-Key: $KEY"
→ {"status":"processing"} … {"status":"done"}

What it costs

A 3-hour recording is 180 minutes — about $1.44 at credit pricing. A 40-recording backlog is roughly $58 to convert in full. $8 per 1,000 minutes — about $0.48 an audio hour — and credits never expire.

Why this needs an API rather than a person

This only works unattended. A human uploading files one at a time is the reason the backlog exists.

Build it on