Captions from a video's speech: $0.20 a job, no SRT needed

Send a finished video with speech to POST /v1/video-captions and Sume transcribes it and burns the captions, at $0.20 per accepted job for videos up to 60 seconds. You do not supply an SRT; the API does not accept SRT uploads. If the clip has no speech it fails with caption_no_speech. See Video captions.

Three ways to get the words

The request

video_url must be a public HTTPS URL Sume can fetch; private, signed and localhost URLs are rejected. Style is optional: Latin text defaults to slam, Korean to black-outline.

curl -X POST https://api.sume.com/v1/video-captions \
  -H "Authorization: Bearer $SUME_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: caption-001" \
  -d '{
    "video_url": "https://media.sume.com/artifacts/example/clean.mp4",
    "style": "punch",
    "script_text": "Say hello to the Sume developer platform."
  }'

When alignment fails

With script_text, the speech-to-text timings stay the source of truth and your words are aligned to them. If your script strays from what is said, the job can end with script_alignment_mismatch or script_alignment_failed; the docs recommend simplifying the script or omitting it to burn the transcript as heard.

At $0.20 a job, 50 captioned clips cost $10.00. Restyling an existing result does not change the price, per the docs. Confirm the live rate in GET /v1/catalog.

Restyle without transcribing again

To burn the same video with a different style, send source_caption_id from the first result plus the new style. The speech-to-text step does not run a second time, and the price does not change. Send words with it only to correct the text.

For a silent clip, use cues or segments with text, start and end; no speech is needed. You can send only one of script_text, words, cues and segments.