A 90-second science explainer from stills, narration and captions

Turn six diagram stills and a 1,200-character script into a captioned class explainer for about $0.66: speech at $0.057, a $0.10 render and $0.20 captions.

4 min readSume
All posts

A teacher can turn six diagram stills and a 1,200-character script into a captioned explainer for about $0.66. Text-to-speech is $0.057, a Timeline render for two minutes is $0.20, and captions are $0.20 for a video of 60 seconds or less.

The captions limit matters here. Sume's video captions are priced for videos up to 60 seconds, so a 90-second explainer is best split into two parts of about 45 seconds each, captioned separately.

Cost of the build

Two 45-second halves of one explainer, Sume list rates read 2026-10-04
StepRateCost
Narration, 1,200 characters$0.0475 per 1,000$0.057
Timeline render, two 1-minute jobs$0.10 per minute$0.20
Captions, two jobs$0.20 per job$0.40
Total$0.657

Plan each still

Give each still 12 to 15 seconds for 90 seconds in total. Put one idea on each: the question, the setup, the change, the result. Timeline takes up to 200 video slots, accepts fit choices such as cover or contain, and offers transitions like fade or dissolve of under a second.

Use contain for diagrams so labels are not cropped. Cover would fill the frame and cut the edges off a chart.

Check before you share

Have a colleague read the script for accuracy. A generated voice reads a wrong formula as confidently as a right one. Compare the caption text with the script, since the captions come from speech recognition unless you send your own.

Sources

Related posts

More in Use cases

All Use cases posts

Written by Sume