Feature Requests

Select multiple visuals and manipulate them at once

Since you are moving towards more AI help in our workflow, in the future, if AI could suggest from all the available libraries the right B-Roll for the scenes we are active in, that would be amazing.

For strokes, outlines and fill colours etc. It would be a massive time saver to have saved brand colours or palettes for default colours

Many filler words that are automatically removed result in a harsh cut of the audio. I would like an option that only removes filler words that can be cleanly removed. For example, if the filler word has a pause between the start and end of it.

I often work on trains and in locations with poor connection. It would be great if there was a true capacity for offline editing like other video editing programs. Even if some AI features would not work until wifi is back. The ability to cut and adjust local files would be amasing.

Please make it more user friendly and easy to do voice overs in the platform.

PLease add the ability for Underlord to use user generated layouts. Underlord is also incredibly confident is does things it absolutely did not do. It hallucinates probably more than any other LLM

There are several attractive female avatars with gray hair, but no male avatars with all gray hair. I need male and female avatars to act as AI Co-Hosts for several websites and YouTube channels I'm building. Seniors are a very important and resource-blessed market. They want to see themselves as competent and intelligent.

Updating the video to a new version removes all existing comments. For future projects, we’ll keep a copy of each version to preserve comment history. Visual discrepancies (e.g., darker slides) may occur between Descript playback and the online export.

We use Descript for serial audio projects with many different audio files and various roles in the team. We write our scripts in Google Docs because only producers have access to Descript and use other software for post production and mixing. For scripting, it would be extremely helpful - actually essential - to be able to automatically copy and paste timestamp/timecodes and file names when copying out text passages (e.g. [00:00:01] - [00:00:31] FILE XYZ) so that the passage is immediately found in post-production (analogous to Trint).