Caption 30 product videos with Sume Video captions for $6

Sume Video captions reserve and capture $0.20 per standalone job, so 30 product videos cost $6.00. Per-batch arithmetic and what the job returns.

5 min readSume
All posts

Each accepted standalone Video captions job on Sume reserves and captures $0.20, so captioning 30 product videos costs 30 x $0.20 = $6.00. The price is per video, so it does not change with the number of words on screen.

Batch cost

Captions at $0.20 per video, as of 2026-10-08 (docs.sume.com/models/video-captions)
VideosCost
10$2.00
30$6.00
100$20.00
52 weekly episodes$10.40

Where it fits

Run captions as the last step on a clip, after trim and composition, so you caption the final timing once. Captioning a draft and then trimming it means paying twice. For a catalog of product videos, run the batch once the cuts are approved.

Doing the batch

Each job is its own request. Give each one an Idempotency-Key built from the video id, so a retry does not charge a second time, and record the result URL against the product. A run of 30 is 30 requests, and the total is the sum above.

What is in the price

The $0.20 is the whole charge for a standalone caption job. The docs state that it is reserved and captured for each accepted job. The docs say this price is for videos of 60 seconds at most, so check the live price on GET /v1/catalog before you queue anything longer.

For a team, the useful number is cost per published video. A caption pass of $0.20 added to a product clip that cost a few dollars to make is a small share, and it makes the clip readable with the sound off.

Quality check

Open three of the 30 results and read the text against the audio, especially product names and numbers. If a brand name is wrong, fix your script text and rerun that video only. The captions docs show a script_text field for passing the known words.

Style and timing options

Captions accept style groups such as phrasing with max_words, max_chars, and pause_seconds, and motion with enter and exit seconds. Set these once in a shared config so all 30 videos look alike. A consistent look matters more for a product line than any single choice of font.

Putting numbers on a month

A shop that publishes 3 product videos a week captions 12 a month, which is $2.40. A catalog back-fill of 300 videos is $60.00, one time. These are small beside the cost of making the videos, so the main question is process, not price: who checks the text and when.

Order of work for a catalog

First, list the 30 videos and their final cuts. Second, caption one and review it. Third, queue the other 29 with the same style config. Fourth, spot-check three. This sequence costs the same $6.00 and finds a styling mistake after one $0.20 job, not after 30. If you have to rerun everything, the cost is another $6.00, which is still small but avoidable.

Keep the receipts

Save each job's id next to the product it belongs to. If a caption needs a fix later, you can find the job, correct the script text, and rerun that single video for another $0.20.

Sources

Related posts

More in Media tools

All Media tools posts

Written by Sume