Feature Requests

It would be helpful that when exporting a transcript as an RTF or other formats, to have a setting to make the speaker labels black. Having the colors looks odd. I can select all change change to black but I do like the speaker labels in gray.

Exporting 1.5hr of audio used to take a few minutes, now it takes 15 minutes. What changed? My computer isn't capping out.

I don't use the web version of Descript and the desktop version used to have the export to YouTube functionality, but for a long time it has said that the feature is coming soon. I don't understand why it was removed in the first place. It seems to me the web version and the desktop version should be identical in features. I post 40+ videos per month to YouTube and it's a pain to export to local and then upload to YouTube, not to mention the drain on my physical resources.

Would be great if Descript can generate 3-5 thumbnails based on the script of the video very similar to how it generates titles.

I choose a frame a my thumbnail and it is very blurry -- low resolution. If you can improve the quality of thumbnail resolution so we don't have to create custom images that would be very helpful

RTF transcript export with AI-generated content produces corrupted characters Description: When exporting a transcript as RTF from the Mac desktop app, any AI-generated content (for example, from Add Chapters) that includes smart punctuation results in corrupted characters in standard editors. Steps to Reproduce: 1. Open Descript (Mac desktop). 2. Generate or edit a transcript using an AI tool (e.g., Add Chapters). 3. Export transcript → RTF. 4. Open the exported RTF in macOS TextEdit or Microsoft Word. Expected Result: All punctuation (curly quotes, non-breaking hyphens, en/em dashes) displays correctly. Actual Result: Unicode typography characters appear as garbled text (mojibake), for example: • Curly single quote (‘) → ‘ • Curly single quote (’) → ’ • Non-breaking hyphen (‑) → ‑ ASCII characters (straight quotes, standard hyphens) display correctly. Additional Notes: • Exporting the same transcript as Markdown (.md) or DOCX does not reproduce this issue, indicating a problem in the RTF export encoding or Unicode handling in Descript. • This occurs immediately upon opening the RTF; no copying or pasting is involved. Environment: • Descript Mac desktop app (latest version) • macOS [insert version] • Test editors: TextEdit, Microsoft Word Impact: Breaks transcripts that contain AI-generated content and forces manual encoding fixes, making RTF export unreliable for workflows that require Unicode-safe punctuation.

Please create a way to download multiple compositions with a single choice / click. I use Descript to split up long Q&As into short bits for Youtube and tiktok. Sometimes I'll have up to 60 different compositions. Then I have to go 1 by 1 and click publish, download, select the next composition, publish, download, select the next composition... over and over and over. It seems like an easy fix. If not a feature to do that automatically, then at least shortcuts for 'next composition' and 'download' would accelerate the process immensely

I duplicated a project multiple times,. When I go into the project to record etc. I see that it retains the original name and shows it inside the script but the top page header shows the new name. The bigger problem is that when I export and download these videos locally all these videos are taking he name from the original (one showing inside the script window).... This is definitely a bug in your system. see attachment

It is way way too difficult to identify small clips and segments of a larger video to export them. The suggestion from your help bot was I had to delete the rest of the video, export what is left, then restore it manually. This is really really not sufficient. Your underlord gets the idea - highlights - but does a really bad unreliable job. I mostly want to control this as I go through a script. My prior editing platform, PowerDirector, does this seamlessly. Do I need to go back to it?

It would be handy to be able to export ALL transcripts for a project in the format we want. It's cumbersome having to navigate to each comp to export the transcripts.