When you first upload a video, Descript's AI will prepare your file by automatically leveling the audio, creating a transcript, and detecting its speakers. The transcript allows you to edit the video as if you would edit a text file. All changes you or the AI makes, will be documented on it as well, so you know at all times what's left in the final cut.
Edit for Clarity - Slice Up Your Timeline
First, when using this feature, you can set the Edit for Clarity intensity. I set mine to medium. Descript edits out all filler words and unnecessary sentences that don’t add to the story. A few “ahms” and “likes” and even two whole sentences later, my video became much clearer.
But at first it went too far. For example, as a joke, I mentioned in one sped-up screen recording that I should add drum and bass music underneath to fit the stressful feel, but Edit for Clarity cut it out, which makes sense for clarity. But I'd leave it in for personality and relatability.
So I went back, changed the intensity (you can also manually select what to keep or remove in the transcription), and then I liked it much better: still sounds like me but more professional without the verbal clutter. This feature is incredible and very useful. What surprised me the most was how smooth the edits are, barely noticeable to the audience or even me.
Remove Filler Words
This feature works similarly to Edit for Clarity. It removed 26 filler words in my almost 4-minute video. Unlike Edit for Clarity, which also removes some comments and jokes, Remove Filler Words just takes out unnecessary things like “ums,” “you know”, “okay,” and such.
Studio Sound Feature
Studio Sound didn’t make a huge difference for this video because my audio quality was already good with a microphone. Therefore, I uploaded another clip where I interviewed the founder of Gamma, Grant Lee, with a suboptimal audio. By the way, another thing I appreciate of Descript is that the software detects the different speakers right away, and marks them in your transcription.
The studio sound is automatically set to 100%, which in my case made it sound artificial. Changing it to 30% upgraded it to podcast quality, smooth and clear. All the background noise disappeared, which makes it much easier to focus on our conversation, without sounding metallic, which other tools tend to do.
Remove Retakes: A Huge Time Saver
Remove Retakes is super helpful. If there is one feature that saves half your work and lots of time, it's this! To test it, I uploaded a raw recording of me talking with no edits. Descript advertises this feature with “Cut all but your best takes.” So, I checked if its choice matched mine.
It took the best versions in almost all of the clips, which I really don't understand how it does it. I said it, it's AI magic.
In a few instances it left bits where I paused, preparing for the next take, which I would cut for sure. I assume that detecting these parts is hard for Descript because the transcription doesn't show them. Still, it’s a huge help and reduces editing work so immensly because you don't have to look through all the footage.
Shorten Word Gaps with Descript
This feature is especially useful for TikTok or YouTube creators or any platform where attention spans are low, and viewers don’t like dragged out sentences. You can specify how long or short the gaps between words should be, and Descript will show you all the places it applies. You can then either remove them all at once or jump to each section and decide individually.
This was the least helpful feature for me, as it mainly detected parts between takes, but it would be useful for interviews where the speaker pauses to think a lot or for webinar replay recordings.
Eye Contact: A Cool Descript Cheat Code
This feature is a little strange but amazing. It perfectly fixes your eyes to look into the camera throughout the whole video. But it doesn't do it in a creepy stalker way, it feels very natural and I had to look at both videos next to each other to see the difference. The only time I I can see that something is off is when I'm technically looking up and thinking, but my eyes look straight into the camera. I think I'm the only one who notices this though.
Another incredibly powerful feature, as my setup for recording with a teleprompter is much more complicated if I require a second screen further away. With the eye correction, I can literally just read off my computer screen, from where I also record and correct my eyes later.
My Unfiltered Conclusion on Descript
I really tried to be picky and share everything about Descript. Honestly, it was hard to find something I didn’t like. This video and audio editing app is so useful, especially for content creators who are often in front of the camera.
The editing process with Descript is very simple; you can streamline workflows easily, and it saves loads of time after filming. Without needing to sort bad takes or fix every detail, you can focus on making much more content. Features like studio sound upgrade your output if you can't film in the best settings or you're on the go.
I love the dashboard and find it very intuitive, the editing software is high-quality and many features are nuanced: just toggle or prompt the intensity you want Descript to apply and it will adjust the content output.
The pricing is very reasonable: the Hobbyist plan costs $16 and includes 10 transcription hours, while the Creator plan is $24 and offers 30 transcription hours. Given how valuable your time is, $16 or $24 is a true bargain compared to the cost of spending those hours editing yourself or hiring help.
Update cookies preferences
