Live-selling replay: 12 clips in 3 languages, captions for $7.68
Trim 12 product clips from a live-selling replay and caption them in three languages on Sume for $7.68: $0.04 a clip to prepare, $0.20 per caption job.
Cutting 12 product clips out of a live-selling replay and captioning each in three languages costs $7.68 on Sume: $0.48 to trim, detach and transcribe 12 clips, plus 36 caption jobs at $0.20. Index-Translate's listed free API translates the cue text.
Assumptions: clips of up to 60 seconds, one transcript per clip, three target languages, and a replay of 30 minutes or less, since the trim source cap is 1,800 seconds. Rates are from the Video trim docs, Video captions docs and pricing page, read 2026-10-05.
The bill
Each caption job is flat for a clip up to 60 seconds, so the three languages are most of the cost.
| Step | Count | Rate | Total |
|---|---|---|---|
| Trim a clip out of the replay | 12 | $0.02 per job | $0.24 |
| Detach the clip audio | 12 | $0.01 per job | $0.12 |
| Speech-to-text, one minute each | 12 | $0.01 per minute | $0.12 |
| Translate cues (Index-Translate public API) | 36 | Listed as free | $0.00 |
| Burn translated cues | 36 | $0.20 per job | $7.20 |
| Total | $7.68 |
Why cues and not the live transcript
Speech-to-text returns words[] with timings. Group them into short cues, translate each cue, and send them as cues. The script_text field aligns your script to the speech timings, so it is for the spoken language, not for a translation.
Where to check
Prices and quantities are the risky text in a live sale. Run the digit check from the linked post on every cue before you burn it, and check the first clip in each language by eye.
Sources
Related posts
More in Use cases
- Localize a 30-second holiday ad into 8 languages for about $2.65
Eight voice-over versions of a 30-second holiday ad cost $2.65 in Sume usage: TTS $0.03, Timeline $0.10, captions $0.20 per language, plus $0.01 to transcribe.
- Localize a 30-second product video into 3 languages: swap the spine
Keep one edit and swap only the voice: three Timeline renders ($0.10) plus three caption jobs ($0.20) cost $0.90 on Sume for English, Spanish and French.
- Localize app store screenshot captions: 6 locales x 5 shots, $1.13
Translate the caption on 5 marketing screenshots into 6 locales with Ideogram 4.5 edits on Sume: 30 calls, $1.125 at low, and why real UI gets re-captured.
- Localize one 30-second ad into six languages: Wan 3.0 cost on Sume
Six language versions of a 30 second ad on Wan 3.0 cost $22.50 at 720p or $45.00 at 1080p on Sume. Draft all six at 480p for $11.25 first.
Written by Sume