Descript transcribes a recording and lets you edit the media by editing the text: delete a sentence and the audio and video are cut accordingly, and removing filler words is a single action across a whole project. Around that idea it offers multitrack editing, screen and remote recording, studio sound enhancement that cleans up poor microphones, a synthetic voice trained on your own speech for fixing mistakes without re-recording, automatic clip generation for social formats, and captions. It suits podcasters, course creators, marketing teams and anyone producing talking-head video who finds a conventional timeline slow. The limitations appear at the edges: it is not a replacement for a professional audio workstation or a colour-grading video editor, transcription accuracy drops with strong accents and overlapping speakers, and projects with long, high-resolution footage can become sluggish on modest hardware.
How to use Descript
1Create an account at descript.com and install the desktop application, which is where serious editing happens even though a browser version exists; a free tier with monthly transcription hours lets you test the workflow.
2Import or record your audio, wait for transcription, then correct speaker labels before editing, since fixing them later is tedious.
3Edit by deleting text, and use the filler word removal and shorten word gaps tools on the whole document rather than passage by passage, as those two actions account for most of the time saved.
4Studio sound improves poor recordings noticeably but can sound processed when pushed, so use it at a moderate level.
5For the synthetic voice feature, record the required training script in the same conditions as your normal recordings or corrections will not match.
6Practical notes: transcription hours are the billing unit, and large video projects benefit from proxy media and a machine with adequate memory.
Best for
podcast production
talking-head video
course recording
clip repurposing
removing filler words
Pros and cons
What it does well
Text-based editing is far faster for spoken content.
Studio sound rescues poor recordings.
Screen and remote recording built in.
Automatic captions and social clip generation.
Where it falls short
Not a substitute for professional audio or video suites.
Transcription accuracy drops with accents and crosstalk.
Large projects can slow down on modest hardware.
Descript pricing
Free tierPrices as listed by the maker; check the site for the latest.