Descript changes how video editing works: it transcribes your recording, and you edit the video by editing the text. Delete a sentence in the transcript and it disappears from the video. No razor tools, no hunting through waveforms.
In this guide you will edit a talking-head video or tutorial from raw recording to finished export: cut mistakes, remove “ums” and long pauses, improve the audio and add captions.
Last updated: October 14, 2026. Menus and button names can change, so if something looks different, look for the closest matching option.
Best for: Talking-head videos, tutorials and podcasts
Starting price: Free plan; paid from $16/month billed yearly
Before you start
- A Descript account and the desktop or web app (the Free plan includes 60 media minutes per month)
- A recording: talking head, screen recording, interview or podcast
- Headphones to check audio edits
- About 10 minutes of footage for your first project
The process at a glance
- Import
- Check text
- Cut by text
- Fillers
- Audio
- Visuals
- Export
Step-by-step: edit your first video in Descript
Step 1: Start a project and import your video
Create a new project and drag your video or audio file into the project window. Descript automatically transcribes it, which usually takes a few minutes.
Record straight into Descript (screen and camera) to skip the import step next time.
Step 2: Check the transcript
Skim the transcript and correct any misheard names or technical terms. Corrections improve your captions later.
Step 3: Cut mistakes by deleting text
Highlight a sentence, false start or off-topic section in the transcript and press Delete. The matching video is removed instantly. Move a paragraph by cutting and pasting it, just like in a document.
Deleted media is not gone forever; you can undo or restore it, so cut confidently.
Step 4: Remove filler words and retakes
Open Underlord (Descript’s AI assistant) and use the AI tools to Remove filler words such as “um”, “uh” and “you know”, and Remove retakes to keep only your best attempt at repeated lines. Review the suggested removals before applying them.
Removing every filler word and pause can make speech sound rushed. Keep some natural pauses, especially before key points.
Step 5: Improve the audio with Studio Sound
Apply Studio Sound to reduce background noise and room echo so your voice sounds like it was recorded in a studio. Adjust the intensity if your voice starts to sound processed.
Step 6: Add captions, B-roll and titles
Add captions from the transcript, then place screenshots, B-roll clips and title cards on the scenes where you introduce key points. For tutorials, zoom into the important part of the screen.
Step 7: Publish and export
Click Publish, open the Export tab, choose the file type and resolution, and click Export. On the Free plan, exports are 720p with a watermark; paid plans unlock higher resolution.
Keyboard habits that save time
- Use the transcript like a document: search (Ctrl/Cmd + F) for a phrase to jump straight to that moment.
- Play from any word by clicking it, then press space to play and pause.
- Edit in passes: first content cuts, then fillers, then audio, then visuals. It is faster than fixing everything at once.
Descript plans for this workflow
| Plan | Price (billed yearly) | Good for |
|---|---|---|
| Free | $0 | Testing with 60 media minutes per month |
| Hobbyist | $16/month | Occasional creators |
| Creator | $24/month | Weekly YouTube or podcast |
| Business | $50/month | Teams and heavier AI use |
Monthly billing costs more ($24, $35 and $65). Remove filler words and more AI tools are included from the Hobbyist plan; the Free plan has limited AI use.
More step-by-step guides
- How to make an AI avatar video with Synthesia
- How to translate a video with HeyGen
- How to make a YouTube video with invideo AI
- How to turn a blog post into a video with Pictory
- How to clone your voice with ElevenLabs
- How to turn a long video into Shorts with Munch
- How to turn an image into a video with Runway
- How to create an interactive training video with Colossyan
- How to make a faceless YouTube video with Fliki
Frequently asked questions
Is Descript free?
Yes, Descript has a free plan with 60 media minutes per month and limited AI tools. Exports on the free plan are 720p with a watermark.
How do I remove “um” and “uh” in Descript?
Open Underlord and use the Remove filler words tool, review the suggestions, then apply them. You can also delete individual filler words in the transcript.
Can Descript edit screen recordings?
Yes. You can record your screen and camera in Descript and edit the recording by editing its transcript.
What is Studio Sound?
Studio Sound is Descript’s AI audio enhancement that removes background noise and echo to make your voice sound professionally recorded.
Can I make YouTube Shorts in Descript?
Yes. Duplicate your composition, change the aspect ratio to vertical, and keep only the best 30–60 seconds.
Wrap-up
Descript turns video editing into document editing: import, fix the transcript, delete what you do not need, remove filler words, polish the audio with Studio Sound, add visuals and export. Edit your next video in passes and you will cut editing time dramatically. See our Descript review or the best AI video editors for YouTube.
This guide is independent and is not sponsored by Descript. Some links may be affiliate links, which never change our recommendations. See our Affiliate Disclosure.