Transcribe a video into sentence segments with Sume: options and cost

Send transcribe: true to POST /v1/video-inspect and Sume runs its speech-to-text on the clip's audio, returning text, words[] and an audio_url. Add segmentation: { "mode": "sentence" } to also get gapless sentence segments[], shaped like caption lines. The docs list $0.01 per audio minute.

Details below come from the Video inspect docs, read 2026-09-29.

Which transcript options exist?

What does a request look like?

Setting frames: false skips the stills, so only the transcript half is billed.

curl -X POST https://api.sume.com/v1/video-inspect \
  -H "Authorization: Bearer $SUME_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: inspect-stt-001" \
  -d '{"video_url":"https://media.sume.com/artifacts/artf_demo/talk.mp4","frames":false,"transcribe":true,"language_code":"en","segmentation":{"mode":"sentence"}}'

What goes wrong?

• video_inspect_transcribe_required: you sent language_code, segmentation or duration_seconds without transcribe: true.
• inspect_source_has_no_audio: the clip has no audio track; check probe.has_audio first.
• ffmpeg_fields_rejected: you sent ffmpeg options; the server builds them.

How is the cost worked out?

Only the transcript reserves; probe and stills are unbilled. Omit duration_seconds and one minute is reserved, so a 10-minute clip needs the hint. The docs list $0.01 per audio minute and say to confirm it in GET /v1/catalog.