How to Clean Descript Captions Without Choppy Audio (Full Transcript)

Use Descript’s filler-word tool to remove ums and uhs from the transcript only, keeping natural audio while producing clean captions.
Download Transcript (DOCX)
Speakers
add Add new speaker

[00:00:00] Speaker 1: If you cut every single um and uh, your video sounds terrible.

[00:00:04] Speaker 2: Cutting out the every time I watch the recordings it feels like this abrupt.

[00:00:07] Speaker 1: If you leave them all in, your captions look terrible. Let me show you how to get clean captions in Descript without removing every single filler word. Go to AI tools, click on remove filler words, pick the ones you want gone like um, uh, and even repeated words. Then change the setting from delete to remove from transcript. And that's it. Now you've got natural, authentic sounding audio and clean captions. Um, like how great is that? I'm Katie, a video editor at Descript. If there's something you want to learn how to do better or faster in Descript, let us know in the comments and we'll make you a short.

ai AI Insights
Arow Summary
The speakers explain that removing every filler word from a video makes audio sound unnaturally abrupt, while leaving fillers in makes captions messy. They demonstrate a Descript workflow to create clean captions without harming natural speech by using AI Tools to remove filler words and switching the action from deleting audio to removing those words only from the transcript. This preserves authentic-sounding audio while improving caption readability, and invites viewers to request more Descript tips.
Arow Title
Clean captions in Descript without ruining natural audio
Arow Keywords
Descript Remove
captions Remove
transcription Remove
filler words Remove
um Remove
uh Remove
AI tools Remove
video editing Remove
remove filler words Remove
authentic audio Remove
workflow tip Remove
Arow Key Takeaways
  • Cutting every filler word can make spoken audio sound choppy and unnatural.
  • Leaving filler words in can clutter captions and reduce readability.
  • In Descript, use AI Tools → Remove filler words to target ums, uhs, and repeated words.
  • Switch the setting from deleting to removing from transcript to keep the audio intact.
  • This approach yields natural-sounding audio with cleaner captions.
Arow Sentiments
Positive: Helpful, upbeat instructional tone focused on improving video quality and offering a simple solution; ends with an inviting call for viewer requests.
Arow Enter your query
{{ secondsToHumanTime(time) }}
Back
Forward
{{ Math.round(speed * 100) / 100 }}x
{{ secondsToHumanTime(duration) }}
close
New speaker
Add speaker
close
Edit speaker
Save changes
close
Share Transcript