Media & ContentBeginnerPreview
Descript for Podcast & Video Editing
A hands-on beginner course for editing audio and video in Descript by editing text instead of a waveform. You will record or import, clean the transcript, cut filler words in one pass, record your screen, use AI tools like Studio Sound and Overdub responsibly, and publish a finished podcast or video.
For podcasters, creators, coaches, marketers, and teams who want to edit talking-head video and audio fast without learning a traditional timeline editor like Premiere or Audition.
Course content
Workbook & downloads
Put the course into practice — a printable workbook plus editable templates you can fill in and reuse.
Preview the workbook
This workbook turns the course into a working production system for editing podcasts and video by editing text. You will set up your project and recording, run a transcript-clean and filler-word pass you trust, build video scenes and screen demos, and finish with a final quality review, export, and publish step. Keep the included templates open as you work — they become your permanent recording-setup sheet, episode production tracker, pre-publish checklist, and content-repurposing log so every future episode starts from a proven, repeatable process instead of a blank project.
Getting Started: How Descript Edits With Text
Set up your project, plan, and transcription correctly, and capture clean source audio and video so every later edit is faster and more accurate.
Worksheet: Project and Account Setup Sheet
Fill this in before you import or record, then confirm each value inside Descript. Getting the plan, language, and vocabulary right now prevents wasted transcription minutes and dozens of text corrections later.
- Content type (solo podcast / interview / talking-head video / screen tutorial)
- Descript plan I am on (Free / Hobbyist / Creator / Business)
- Monthly transcription hours I actually need
- Spoken language set before transcription
- Custom vocabulary terms added (guest names, brand names, jargon)
- One project per episode confirmed (Y/N)
- Will I need Studio Sound or Overdub this project? (Y/N)
Worksheet: Clean-Source Recording Sheet
Complete this before you hit record, or before you import existing files. The cleaner the source, the more accurate the transcript and the less you fight Studio Sound later.
- Room (quietest, softest space available)
- Microphone and input device
- Mic distance from mouth (target 15 to 20 cm, slightly off-axis)
- Peak level checked near -12 to -6 dBFS, never 0 (Y/N)
- Each remote guest on a separate track? (Y/N)
- Notifications and noisy devices silenced (Y/N)
- Source files are highest-quality originals, not re-exports (Y/N)
Exercise: Edit-One-Sentence Drill
Before making anything real, prove the transcript model to yourself on a 60-second test recording until the core idea is obvious.
- Record or import a one-minute clip and wait for the transcript to generate.
- Highlight one rambling sentence in the transcript and press Delete — watch the audio and video disappear with it.
- Select a different sentence and use Ignore instead; confirm it strikes through, skips on playback, and is reversible.
- Cut a sentence from the end and paste it earlier, then play back to confirm the media moved with the words.
Checklist: Ready-to-Edit Sanity Check
- Correct spoken language selected before transcription ran
- Custom vocabulary loaded with names, brands, and jargon
- Source audio is clean, close, and consistent (or imported at full quality)
- Each speaker on a separate track where the recording allowed it
- One project created per episode or video
- Transcript generated and skimmed once for obvious errors
Editing the Transcript: The Core Workflow
Cut and rearrange by editing words, run a filler-word and pacing pass that still sounds human, and balance multiple voices with sensible levels and Studio Sound.
Exercise: Restructure-by-Text Exercise
Practice the moves you will use on every project by restructuring a real recording purely in the transcript, never opening the timeline.
- Delete three weak or rambling sentences from the opening two minutes.
- Find your single strongest moment and cut-and-paste it nearer the start.
- Correct two misheard words in the transcript and proofread the surrounding lines for captions.
- Use Ignore on a section you are unsure about rather than deleting it, so it stays recoverable.
Worksheet: Filler-Word and Pacing Pass Log
Record your decisions for the filler and pacing pass so it stays consistent and human across every episode. Fill it in as you run the pass.
- Total fillers detected
- Fillers removed in bulk (um, uh)
- Connectors reviewed individually (you know, like, so)
- Approx. share of natural connectors deliberately kept
- Gap-shortening threshold used (e.g. trim silence over 1.0s)
- Spots where clustered cuts sounded choppy and were undone
- Runtime before pass / after pass
Exercise: Multi-Speaker Balance Drill
On a two-voice recording, practice per-track control and Studio Sound so neither speaker is too loud, too quiet, or too roomy.
- Label each track with the speaker's name.
- Raise the quieter speaker and lower the louder one until they feel even.
- Mute a cough or crosstalk on just one track without touching the other.
- Apply Studio Sound to only the roomier track at medium strength and listen for artifacts.
Checklist: Core-Edit Done Check
- Content restructured for a strong opening using text edits
- Transcript corrected and proofread for accuracy
- Filler pass run, with a light scatter of natural connectors kept
- Long gaps shortened without making speech sound clipped
- Speaker levels balanced to a consistent loudness
- Studio Sound applied per track only where needed, with no artifacts
Video, Screen Recording, and AI Tools
Build video with scenes, layers, B-roll, and proofed captions, capture a clean screen demo, and use Overdub and other AI tools responsibly and with consent.
Worksheet: Video Scene and Layer Plan
Sketch your video before you build it so every layer earns its place and nothing covers what matters. Fill one row of fields per major scene.
- Scene name and what is happening
- Base layer (webcam full-frame / corner / screen recording)
- B-roll or image layers and the sentence each supports
- Text or lower-third titles (names, key points)
- Captions enabled and positioned in the safe zone (Y/N)
- Layout for this scene (full-frame / split / picture-in-picture)
Exercise: Clean Screen-Demo Exercise
Record a short, watchable screen demo by preparing the screen first and editing the words after.
- Close notifications and private tabs, and raise system zoom so text is legible at video resolution.
- Narrate a two-minute demo from a bullet outline rather than reading word for word.
- Edit it in the transcript: delete dead clicks and mistakes, then run a filler pass.
- Add your webcam as a corner layer for the intro and hide it during dense screen work.
Worksheet: Overdub Consent and Use Record
Document responsible voice-cloning use before you rely on Overdub. Complete this whenever you create or use a custom voice.
- Voice owner (must be you or someone who gave permission)
- Vocal consent recording completed in Descript (Y/N)
- Written permission on file if cloning someone else (Y/N)
- Corrections made by Overdub (word or short phrase each)
- Any passage longer than ~2 sentences re-recorded instead (Y/N)
- Collaborators informed where synthetic audio appears (Y/N)
Checklist: Video and AI Review Check
- Every layer clarifies the words underneath it, nothing is filler
- Captions generated, styled, proofread, and inside the safe zone
- Screen demo recorded on a tidy desktop with legible text
- Overdub used only for short fixes, with consent completed
- Underlord drafts (titles, chapters, notes, edits) reviewed and corrected
- Eye Contact or Green Screen, if used, checked for natural results
Polish, Export, and Publish
Run a structured final quality pass, export at the right settings, publish a podcast with chapters and show notes, and repurpose the recording into many assets.
Checklist: Pre-Export Quality Checklist
- Full uninterrupted listen on representative headphones or speakers
- Transcript and on-screen captions proofread, names spelled right
- Speaker levels balanced and overall loudness consistent
- Studio Sound checked for artifacts; filler pass checked for choppiness
- Intro, outro, music beds, B-roll, and titles correct and timed
- All on-screen text and lower-thirds spell-checked
Worksheet: Export Settings Decision Sheet
Decide and record your export settings per destination so they are consistent every time. Confirm watermark status and resolution caps on your current plan.
- Destination (podcast host / YouTube / Shorts-Reels-TikTok)
- Audio format and bitrate (WAV master, or MP3 192 to 320 kbps)
- Mono or stereo (mono fine for single voice)
- Video format, resolution, frame rate (e.g. MP4 1080p 30fps)
- Vertical 1080x1920 for short-form? (Y/N)
- Loudness target (around -16 LUFS, true peak below ~-1 dBTP)
- Watermark removed and resolution cap acceptable on my plan (Y/N)
Exercise: Publish-and-Repurpose Exercise
Take one finished episode all the way to published and then squeeze multiple assets from it.
- Export the audio and upload it to your host (e.g. Buzzsprout, Transistor, Spotify for Podcasters), then write a title and show notes.
- Have Underlord draft chapters and show notes, then edit them for accuracy before publishing.
- Highlight one strong 30 to 60 second moment, switch it to vertical, add captions, and export a clip.
- Create one more asset — an audiogram or quote graphic — from the same project.
Checklist: Shipped-and-Repurposed Check
- Episode exported at the correct settings for its destination
- Uploaded to host with edited, fact-checked show notes
- Chapter markers added where helpful
- Transcript published or attached for accessibility and SEO
- At least two short clips cut for social from the same recording
- Repurposing log updated with what was published where
Your Action Plan
- Confirm your Descript plan covers your monthly transcription hours, set the spoken language, and load a custom vocabulary of names, brands, and jargon.
- Record or import clean, close, consistent source audio with each speaker on a separate track wherever possible.
- Let Descript transcribe, then proofread and correct the transcript so it is accurate for editing and captions.
- Restructure the piece by editing text — delete weak lines, move your strongest moment near the start.
- Run Remove Filler Words on um and uh, review connectors individually, and shorten long gaps while keeping speech human.
- Balance speaker levels and apply Studio Sound per track only where the room or noise demands it.
- If it is video, build scenes and layers, add timed B-roll and proofed captions, and record any screen demo on a tidy desktop.
- Use Overdub only for short fixes with consent completed, and review every Underlord-drafted title, chapter, or note.
- Do the full final listen and run the pre-export checklist, then export at the correct settings for the destination.
- Publish with edited show notes and chapters, then cut at least two clips and one audiogram to repurpose the recording.
Pairs well with
Courses members commonly take alongside this one.
Flagship CoursePreview
Freelance Business Foundations: Position, Price, Sell, and Deliver High-Value Services
Freelancing · Beginner · 16h
Self-pacedPreview
Client GrowthPreview
Freelance Client Acquisition: Outreach, Leads, Referrals, and Deal Flow
Freelancing · Beginner · 15h 30m
Self-pacedPreview
Sales SystemPreview
Freelance Sales & Proposals: Discovery Calls, Scoping, Objections, and Closing
Freelancing · Intermediate · 16h
Self-pacedPreview