On the next page, you will choose a transcript. You have three options:
Use YouTube's autogenerated transcript. This gives you word-level alignment, but no punctuation. It can vary from "pretty good!" to "wildly inaccurate!"
Use a transcript that was manually uploaded to YouTube. These can be extremely accurate, but sometimes creators upload captions that are even worse than the autogenerated data.
Provide your own SRT or VTT captions file. You can create one automatically with Descript (see this how-to), or manually with VTT Creator.
Create the Video Digest
After successfully uploading a video and transcript, you can start creating the video digest via our editing interface: The "Editing Instructions" pane in the bottom right lists the actions that you can perform. Here's an example of a few common actions:
Segment the video into chapter elements
You can create a new chapter element by clicking on the transcript while holding the ctrl+alt (cmd+alt on a mac) keys:
Segment the video into section elements
You can create new section elements by clicking on the transcript while holding the ctrl (cmd on a mac) key:
Edit text elements
You can edit the text elements by clicking on them and changing the text:
Write section summaries
You can write section summaries by clicking on the section boxes and typing a summary into them:
You can change the keyframe elements by navigating the video to the desired location and clicking the camera icon:
Preview / Save / Publish / Share the Video Digest
You can use the toolbox in the bottom right corner to preview, save, and publish the final video digest,