Quick Start

This guide covers every Descript feature:
- Getting Started — Create account and basic setup
- How to Use Text-Based Editing — Cut audio and video files by deleting words in a transcript
- How to Use AI Voice Cloning — Fix a flubbed line by typing it in your own voice
- How to Use Studio Sound — Remove background noise and echo from any audio file
- How to Use Filler Word Removal — Delete every um and uh across a recording with one click
- How to Use Multitrack Editing and Collaboration — Edit separate audio tracks for each guest and share the project with your team
- How to Use Screen Recorder — Capture screen recordings and webcam video without a second app
- How to Use Overdubbing — Patch a mistake mid sentence without re-recording the whole clip
- How to Use Automatic Transcription — Automatically transcribe audio and video content in 22 languages
- How to Use AI Green Screen — Remove your background with no physical green screen
Time needed: 5 minutes per feature
Also in this guide: Pro Tips | Common Mistakes | Troubleshooting | Pricing | Alternatives
Why Trust This Guide
I have used Descript for two years across podcasts, client tutorials, and YouTube uploads. Every step here comes from real projects, not a press kit.

Descript is a video editor and audio editor built around a transcript.
You edit audio and video by editing words, which is why beginners pick it up quickly.
It is a powerful tool for producing high quality content in one audio video workflow.
Most people stop at cutting text and never touch the advanced features.
This guide walks through all nine, step by step, with screenshots and pro tips.
The goal is a faster editing process that still protects your creative vision.
Descript Tutorial
This How to Use Descript tutorial covers setup first, then every feature, from text based editing to AI Green Screen.

Descript
Edit video by editing text. Descript transcribes your files, strips background noise with Studio Sound, and cuts filler words in one click. Start on the free plan, no card needed.
Getting Started with Descript
Setup takes about three minutes and only happens once.
Here is how Descript works in a real project:
Now let’s walk through each step.
Step 1: Create Your Account
Go to the Descript website and click Sign Up.
Email, Google, or single sign-on all work for a new account.
✓ Checkpoint: Check your inbox for a confirmation email.
Step 2: Install the Desktop App
Descript works in Chrome, but the Mac and Windows apps handle large video files better.
There is no mobile app, so plan to edit at a desk.
Here is what the workspace looks like:

✓ Checkpoint: You see the script editor on the right and the scene rail below.
Step 3: Create Your First Project
Click New Project, then upload an audio or video file to start.
Projects hold big productions, and compositions hold alternate cuts of the same recording.
✅ Done: You’re ready to use any feature below.
How to Use Descript Text-Based Editor
Text-Based Editing lets you cut audio and video files by deleting words in a transcript.
Text based editing is the reason most people switch to Descript. Delete a sentence in the script editor and the matching footage disappears from the timeline.
Step 1: Open Your Transcript
Import a video file and wait for the transcript to load.
The script editor opens beside your timeline, and it reads like a word document.
Step 2: Delete the Bad Takes
Highlight the words you want gone and press delete.
Here’s what this looks like:

✓ Checkpoint: Deleted words show as a strikethrough line, and the video skips them.
Step 3: Play It Back
Press space to review the cut before you export anything.
✅ Result: Your first text based editing pass is done in minutes, not hours.
💡 Pro Tip: Enable Ignore Deleted Text to hide old lines and keep the script clean while editing.
How to Use Descript AI Voice Cloning
AI Voice Cloning lets you fix a flubbed line by typing it in your own voice.
AI voice cloning trains on a short recording of your own voice. Descript then speaks new words for you when a re-record is not practical.
Step 1: Train Your Voice
Record the consent script Descript gives you, roughly ten minutes of clean audio.
Step 2: Type the Fix
Click into the transcript and type the corrected words.
Here’s what this looks like:

✓ Checkpoint: The new words appear in the waveform with a small AI badge.
Step 3: Blend the Audio
Adjust pacing so the generated audio matches the original audio around it.
✅ Result: One typed line replaces a full re-record session.
💡 Pro Tip: Train voice cloning in the same room and mic you always use, or the patch will stand out.
How to Use Descript Studio Sound
Studio Sound lets you remove background noise and echo from any audio file.
Studio Sound is the fastest way to make a laptop mic sound close to a booth. It strips room echo, hiss, and background noise in one pass.
Step 1: Select the Track
Click the audio layer you want to clean up.
Step 2: Toggle Studio Sound
Turn it on in the right hand panel and set the strength slider.
Here’s what this looks like:

✓ Checkpoint: Room echo drops away and the voice sits forward in the mix.
Step 3: Compare Both Versions
Toggle the effect off and on to hear the difference.
✅ Result: Rough audio files now sound like planned studio takes.
💡 Pro Tip: Keep the slider near 70 percent, since full strength can hollow out a deep voice.
How to Use Descript Filler Word Remover
Filler Word Removal lets you delete every um and uh across a recording with one click.
Descript tags filler words in the transcript automatically. One click clears them from the whole project instead of hunting through a waveform.
Step 1: Open the Edit Menu
Choose Remove Filler Words from the editing tools.
Step 2: Review the Highlights
Descript underlines each filler word it found in the script.
Here’s what this looks like:

✓ Checkpoint: The word count drops and the pacing tightens noticeably.
Step 3: Remove Them All
Confirm the removal, then spot check two or three sentences.
✅ Result: A rambling take turns into a clean cut without manual trimming.
💡 Pro Tip: Leave a few natural pauses in podcasts, because zero filler words can sound robotic.
How to Use Descript Multitrack Editor and Collaborator
Multitrack Editing and Collaboration lets you edit separate audio tracks for each guest and share the project with your team.
Every speaker keeps separate audio tracks, so you can duck one mic without touching another. Teammates can open the same project and comment in real time.
Step 1: Import Every Track
Drag and drop each guest recording into one composition.
Step 2: Label the Speakers
Assign names so the transcript splits cleanly by person.
Here’s what this looks like:

✓ Checkpoint: Each speaker shows on their own lane in the timeline.
Step 3: Invite Your Team
Share a link and set edit or comment access for each person.
✅ Result: Multi guest video projects stay organised from raw files to final cut.
💡 Pro Tip: Business accounts add single sign-on, which is worth it once more than five editors share the account.
How to Use Descript Screen Recorder
Screen Recorder lets you capture screen recordings and webcam video without a second app.
The built in screen recorder captures your display, mic, and camera together. Each source lands as its own layer for editing.
Step 1: Open the Record Panel
Click New Recording and pick screen, camera, or both.
Step 2: Hit the Record Button
Press the record button and run through your demo.
Here’s what this looks like:

✓ Checkpoint: Your screen recording appears in the project with a ready transcript.
Step 3: Stop and Transcribe
End the video recording and Descript starts the transcript on its own.
✅ Result: A tutorial goes from capture to editable script in under two minutes.
💡 Pro Tip: Record at 1080p for tutorials, since text on screen turns mushy at lower settings.
How to Use Descript Overdubbing
Overdubbing lets you patch a mistake mid sentence without re-recording the whole clip.
Overdub is the repair tool that sits on top of voice cloning. It swaps a wrong number or name inside a finished audio file, and it doubles as text to speech in your own voice.
Step 1: Find the Error
Search the transcript for the wrong word.
Step 2: Type the Replacement
Select Overdub and type what you meant to say.
✓ Checkpoint: The corrected phrase plays in your cloned voice at matching volume.
Step 3: Check the Seams
Listen to the sentence twice and nudge the timing if needed.
✅ Result: One wrong stat no longer costs you a full re-record.
💡 Pro Tip: Overdub short phrases only, because long generated passages lose natural emphasis.
How to Use Descript Automatic Transcriptor
Automatic Transcription lets you automatically transcribe audio and video content in 22 languages.
Descript will automatically transcribe anything you upload, including German, Spanish, and French. Accuracy is strong, though names and heavy accents still need a read through.
Step 1: Upload Your File
Drop in an audio or video file from your drive.
Step 2: Pick the Language
Choose from 22 supported languages before processing starts.
Here’s what this looks like:

✓ Checkpoint: Timestamped text appears next to the waveform within a minute or two.
Step 3: Fix the Names
Use Correct Text to fix spellings the AI tool missed.
✅ Result: You get a searchable script you can also publish as show notes.
💡 Pro Tip: Add proper nouns to the vocabulary list first, and transcription accuracy jumps on the next file.
How to Edit Descript AI Green Screen
AI Green Screen lets you remove your background with no physical green screen.
AI Green Screen cuts you out of a messy room in one click. Pair it with Eye Contact and a talking head clip starts to look deliberate.
Step 1: Select the Camera Layer
Click your webcam clip in the timeline.
Step 2: Apply Green Screen
Turn on the effect and choose a colour, image, or blur.
✓ Checkpoint: Your background swaps instantly while you stay sharp in frame.
Step 3: Refine the Edges
Adjust the threshold until hair edges stop flickering.
✅ Result: A kitchen table shoot passes for a planned set.
💡 Pro Tip: Even lighting on your face beats any slider, so face a window before you hit record.
Descript Pro Tips and Shortcuts
After two years of daily editing, these are the shortcuts I actually use.
Keyboard Shortcuts
| Action | Shortcut |
|---|---|
| Play or pause | Space |
| Split at cursor | Cmd / Ctrl + E |
| Ignore selected text | Cmd / Ctrl + Shift + E |
| Start a new recording | Cmd / Ctrl + Shift + 2 |
Hidden Features Most People Miss
- Underlord: The assistant editor takes plain language prompts, drafts social posts, and generates B roll. Descript markets it as apocalyptic power at inhuman speed, which is marketing, but it is quick.
- Eye Contact: Redirects your gaze to camera when you were reading a script, so a scripted take still feels live.
- Timeline export: Send an XML to Final Cut Pro or Premiere when a client wants colour grading elsewhere.
- Animated captions: Pull captions and overlays straight from the transcript to lift video content on social.
Descript Common Mistakes to Avoid
Mistake #1: Editing before you fix the transcript
❌ Wrong: Cutting clips while names are still spelled wrong in the script.
✅ Right: Correct the transcript first. Every later edit points at the right words, and your captions ship clean.
Mistake #2: Running Studio Sound on everything
❌ Wrong: Applying full strength Studio Sound to music beds and room tone.
✅ Right: Use it on voice only. Music processed as speech turns thin and warbly fast.
Mistake #3: Exporting before checking the free plan limits
❌ Wrong: Finishing a long video, then finding the export is capped at 720p.
✅ Right: Check your tier first. The free plan gives one watermark free video export per month at 720p.
Descript Troubleshooting
Problem: Transcription accuracy drops on interviews
Cause: Two people on one mic, heavy accents, or background noise in the room.
Fix: Record separate audio tracks per speaker, run Studio Sound to remove background noise, then transcribe again.
Problem: Voice cloning sounds flat
Cause: Training audio was recorded on a different mic or in a louder room.
Fix: Retrain with ten clean minutes from your usual setup, and keep overdubbed patches under one sentence.
Problem: Exports fail or stall
Cause: A 4K video recording plus AI effects can exhaust local memory during rendering.
Fix: Close other apps, export at 1080p, or split long video projects into shorter compositions.
📌 Note: If none of these fix your issue, contact Descript support through the in app chat.
What is Descript?
Descript is an AI powered editor for audio and video content, built on an editable transcript.
Think of it as a word document that happens to control your timeline.
Watch this quick overview:
It includes these key features:
- Text-Based Editing: Cut audio and video files by deleting words in a transcript.
- AI Voice Cloning: Fix a flubbed line by typing it in your own voice.
- Studio Sound: Remove background noise and echo from any audio file.
- Filler Word Removal: Delete every um and uh across a recording with one click.
- Multitrack Editing and Collaboration: Edit separate audio tracks for each guest and share the project with your team.
- Screen Recorder: Capture screen recordings and webcam video without a second app.
- Overdubbing: Patch a mistake mid sentence without re-recording the whole clip.
- Automatic Transcription: Automatically transcribe audio and video content in 22 languages.
- AI Green Screen: Remove your background with no physical green screen.
Editing is non destructive, so your original files stay untouched while you experiment.
Finished work exports to YouTube, Spotify, or any podcast host, including Hello Audio.
For a full review, see our Descript review.

Descript works for solo creators, but teams get the most out of it.
Descript Pricing
Here’s what Descript costs in 2026:
| Plan | Price | Best For |
|---|---|---|
| Free | $0 | Testing Descript with one transcription hour a month |
| Hobbyist | $16 | Personal use and short clips at 1080p |
| Creator | $24 | Weekly podcasts, stock media, and 4K exports |
| Business | $50 | Teams needing single sign-on and dubbing |
| Enterprise | Custom | Large orgs with dedicated support |
Free trial: Yes. The free plan runs forever with one transcription hour each month.
Money-back guarantee: No formal refund window, but a yearly paid subscription saves you 20 percent.
Paid tiers unlock unlimited use of AI effects, 4K exports, and the stock media library.

💰 Best Value: The Creator plan, because 4K exports and stock media cover almost every solo creator.
Descript vs Alternatives
How does Descript compare? Here’s the competitive landscape:
| Tool | Best For | Price | Rating |
|---|---|---|---|
| Descript | Text based editing | $16/mo | ⭐ 4.5 |
| CapCut | Short form clips | $8/mo | ⭐ 4.5 |
| Filmora | Effects and scenes | $20/mo | ⭐ 4.4 |
| VEED | Browser subtitles | $19/mo | ⭐ 4.4 |
| Animoto | Marketing slideshows | $16/mo | ⭐ 4.2 |
| InVideo | AI video generation | $28/mo | ⭐ 4.3 |
| Gling AI | YouTube cleanup | $15/mo | ⭐ 4.2 |
Quick picks:
- Best overall: Descript — one tool for recording, transcripts, and editing.
- Best budget: CapCut — free for most short clips.
- Best for beginners: Descript — if you can edit a document, you can edit video.
- Best for effects: Filmora — deeper transitions and titles.
🎯 Descript Alternatives
Looking for Descript alternatives? Here are the top options:
- 🚀 CapCut: Free mobile first video editor with fast templates, though it has no transcript driven editing or voice cloning.
- 🎨 Filmora: Timeline video editing with deep effects and titles, better suited to creators who think in scenes, not scripts.
- ⚡ VEED: Browser based editor with strong subtitles and quick social exports, no desktop install needed.
- 💼 Animoto: Drag and drop slideshow maker for marketing videos built from stock media and photos.
- 🧠 InVideo: AI powered generator that turns a prompt or script into a finished video with stock footage.
- 🎯 Gling AI: Specialised tool that removes filler words and bad takes from YouTube footage automatically.
For the full list, see our Descript alternatives guide.
⚔️ Descript Compared
Here’s how Descript stacks up against each competitor:
- Descript vs CapCut: Descript wins on text based editing and podcasts. CapCut wins on free short form clips for social.
- Descript vs Filmora: Filmora offers richer visual effects. Descript is faster for talking head video content and long interviews.
- Descript vs VEED: VEED is simpler for one off subtitles. Descript gives you full control over multitrack audio and video.
- Descript vs Animoto: Animoto suits promo videos from images. Descript suits spoken audio and video editing at length.
- Descript vs InVideo: InVideo creates video from scratch. Descript edits your own recording with more precision.
- Descript vs Gling AI: Gling AI does one job cheaply. Descript offers the same cleanup plus recording, transcripts, and export.
Start Using Descript Now
You learned how to use every major Descript feature:
- ✅ Text-Based Editing
- ✅ AI Voice Cloning
- ✅ Studio Sound
- ✅ Filler Word Removal
- ✅ Multitrack Editing and Collaboration
- ✅ Screen Recorder
- ✅ Overdubbing
- ✅ Automatic Transcription
- ✅ AI Green Screen
Next step: Pick one feature and try it now.
Most people start with text based editing, and it takes less than five minutes.
Frequently Asked Questions
What is Descript and how does it work?
Descript is an AI powered video and audio editor. It transcribes your file, then edits the media when you edit the text, so cutting words cuts footage.
How do you transcribe on Descript?
Upload an audio or video file and Descript will automatically transcribe it in one of 22 languages. Then use Correct Text to fix names and accents.
How to use Descript for a podcast?
Record each guest on separate audio tracks, run Studio Sound, remove filler words, then export the audio file to Spotify or your podcast host.
Is Descript good for transcription?
Yes, accuracy is high on clean audio. It still trips on names, crosstalk, and strong accents, so read the script before publishing.
What is Descript used for?
People use Descript for podcasts, YouTube videos, screen recordings, and course clips. It handles recording, editing, captions, and export in one place.













