How to use Descript to edit video and podcasts like a doc
Step-by-step Descript guide for beginners: transcribe, cut by deleting words, remove filler words, add captions and export, plus when Descript is (and isn't) the right tool.
Descript turns your recording into a transcript. Delete words in the text and the matching video/audio is cut. It's ideal for talking-head videos, podcasts, interviews and screen recordings.
Import and transcribe
Create a project, drag in your video or audio, and let it transcribe. Fix speaker labels so cuts land on the right person.
Rough cut by editing text
Read the transcript like a doc. Highlight and delete flubbed takes, tangents and repeats. Play back each cut; adjust if a breath or word is clipped.
Clean up filler words and silences
Use the built-in tools to find filler words ("um", "uh") and long gaps, and review before removing all. Removing every one can sound robotic, so keep a few.
Improve audio
Try the studio-sound/audio-enhancement feature on noisy recordings, and compare before/after: it works best on voice, less well on music.
Captions, titles and clips
Add captions from the transcript (great for social), then ask an assistant for titles and descriptions:
Here's the transcript of my video. Suggest 5 titles under 60 characters, a 150-word description, and 3 moments that would make good 30-60 second clips with timestamps.
[paste transcript]When to use something else
For heavy effects, color grading or music videos, a traditional editor is better. For AI-presenter videos with no camera, look at Synthesia or HeyGen.
Quick tips on this topic
⚡ Delete words, cut video
In Descript, deleting a word in the transcript cuts it from the video. Edit like a doc.
⚡ Don't remove every 'um'
Stripping all filler words can sound robotic. Review and keep a few.
Swipe 50+ quick AI tips →One tip per screen, 10 seconds each