Feature Requests

Sometimes when working with Zoom meeting recordings, a participant's web cam may have extra delay and is behind in terms of sync with the audio track. In a meeting with multiple participants, each participant may have a different time delay between audio and video sync, so the issue must be fixed accordingly to each speaker in the video file. I would like a way--perhaps with AI--to automatically detect bad lip sync and to automatically fix it by any number of ways, shifting audio, time stretching if there is a clocking issue, or some other novel way.

Good morning, I don't plan on using your service in the following months. Would it be possible to reimburse me for the payment made this month, please? As you can see in my profile, I have not reused your services since February. Thanks for your understanding.

Open the API access to give paying users more options outside of Descript and help Descript get more visibility with different use cases (e.g. integrate into applications like GPT connects into other tools and gain even more popularity). Win-win 🙌

The Underload feauture should be able to keep a history of prompts that were inputed in the past, so that when users come back to the project, they remember where they left off

I love that it can create chapters for me but it usually makes too many. And some of them are literally a sentence long, so not that helpful

Please consider expanding the range of OpenRouter models available as AI backend options. For example, recently released free models such as Healer Alpha, which have demonstrated strong performance in video understanding and agentic tasks, would be valuable additions.

I'm trying to build an automatic editing flow that requires the Underlord to have the ability to insert gap clips at transition points in the show. However, It seems to be unable to add gap clips precisely where existing markers are and so it's trying to do a workaround that estimates the placement of where it will insert an empty scene but it's not precise enough and is not going well. Chatting with Underlord, it says: Looking at the tools I have access to: add_scene_boundaries — inserts empty scenes, but snaps to word boundaries and can't target exact timestamps at existing scene boundaries edit_narration — edits spoken content (ignore, delete, replace, insert TTS), but has no concept of inserting silence. INSERT creates TTS speech, REPLACE regenerates audio, and neither produces a clean gap clip shorten_word_gaps — can reduce pauses, but can't add or lengthen them insert_track — inserts media overlays, but can't insert silence into the script track itself None of these are the "Add Gap Clip" function.

Please use Underlord's features for German videos as well. Record German videos. Otherwise it makes little sense without the functions

Please add a feature like what Wisecut does and Timebolt but automatic and much better

Laughter in a recording is not transcribed, and shows up as "silence." Thus bulk remove of silence removes laughter, and other non-word communication (gasps, sighs, etc.). Ideally transcription would tag common exclamations as Laughter, Sigh, etc. Barring that, it would be nice if it could distinguish between actual silence, and non-word sounds in transcription so that shorten silences wouldn't cut out things like that, or could be at least be configured not to. It seems like the "tagging" functionality would be a good match for this, however that does not work on Multitrack, and this feature would be needed in Multitrack as well.