EchoVideo

From an instructor point of view, would it be possible to generate polling questions to the video based on the transcript and time stamp? Or manually add the questions for users who want more control? I've seen this being done in another platform.

Detect and remove silent or non-instructional segments automatically. Option to review and approve suggested cuts. Automatically generate chapters based on key topic shifts or slide changes. Use AI to highlight important moments or frequently rewatched sections. Use AI to clean up audio—reduce noise, amplify instructor voice, and improve overall clarity.

We recently realized that the AskEcho functionality is only available in folders. I expected AskEcho functionality to be available inside of embedded content, especially content embedded with LTI 1.3. We would really like to see this available in course sections. This would be particularly useful before the Folders: Phase 4 migration tool is available to migrate from course sections to folders. We like the 80% threshold for AskEcho Content (Summary and FAQs) not displaying till the recording has been viewed but I think it should be higher like 90% or make it only show once the student has viewed the recording at 99%. I think that being at 80% and automatically displaying students, they would catch on and not view the recorded lesson in its entirety and skip ahead play it and get the AskEcho Summary and FAQs. Or does AskEcho already have that functionality that the video must be played all the way through to the threshold with no skipping? We also had some concerns about students copying the AskEcho content and using it to cheat. I tested this with our test student and clicked on the “Save to my Notes” it saves to my Notes panel successfully. However, if I wanted to copy/paste the content I could. This will be a caveat for some faculty members, I know a generous amount of them already don’t like that students can download a transcript for fear of using AI tools to cheat. Yes, they can control what is displayed in the AskEcho Content so it could be a teaching moment for our team being it would be up to the faculty member to review the AskEcho content that was generated. Regarding the links, maybe others would like to AskEcho Content to be visible in links, (public links are turned off at our institution) but perhaps adding a toggle that admins can control.

Getting good quality audio from all captures is challenging. Can AI speech enhancement models be used to generate higher quality audio tracks? This would in turn improve the accuracy of captions. Other examples of this are well documented, ie https://podcast.adobe.com/en/enhance