Loading tutorials…
Loading tutorials…
If your Descript exports have sync drift, blurry video, inconsistent audio levels, or fail to render, the cause is almost always one of 7 specific issues. This walks each one with the exact fix.
Who this is forAnyone whose Descript exports have visible quality issues. Common symptoms: A/V sync drift, blurry video, audio level mismatch between speakers, exports stuck rendering, or files that won't play.
What you'll need
Step 1
Play the exported file in QuickTime / VLC / your destination platform. Note exactly what's wrong. Match to the fix below.
Open the export in QuickTime (Mac), Media Player (Windows), or VLC. Watch the first 60-120 seconds.
Categorize the issue:
— A/V out of sync (audio drifts ahead of or behind video over time) → Step 2
— Blurry / soft video despite high bitrate → Step 3
— Audio levels inconsistent (one speaker much louder than another) → Step 4
— Audio robotic or compressed → Step 5
— Export stuck at X% or fails → Step 6
— File won't play on destination platform → Step 7
Multiple issues at once: fix in order above. A/V sync issues compound when other audio issues are present during diagnosis.
Step 2
Source files mismatched. Re-import source files. Verify in project. Re-export. If persistent: render at uniform frame rate (29.97 → 30 fps consistent).
A/V drift in Descript exports usually comes from one of three causes:
Cause 1: source files have inconsistent frame rates. Example: webcam recorded at 29.97 fps, screen recording at 60 fps, podcast guest video at 24 fps. Descript struggles to align.
Fix 1: in the project, verify Frame Rate in Project Settings → set to 30 fps (or your target). Re-render.
Cause 2: source audio file is variable bit rate (VBR). Causes sync drift over long files.
Fix 2: convert source audio to constant bit rate (CBR) MP3/WAV in Audacity or Adobe Audition before re-importing. Re-import the CBR version. Re-export.
Cause 3: corrupted source file. Symptom: drift starts at a specific timestamp and increases.
Fix 3: re-record (if possible) or trim around the corrupted section.
If all else fails: export audio and video separately, then re-mux in Premiere or FFmpeg with explicit sync. Last resort.
Step 3
Bitrate too low. Set export bitrate to 12-15 Mbps for 1080p, 35-50 Mbps for 4K. Use H.264 codec for compatibility, H.265 (HEVC) for smaller files.
Blurry exports despite high source quality = export bitrate too low or wrong codec.
Publish → Export → Advanced. Set:
— 1080p: 8-12 Mbps minimum, 15 Mbps for sharper quality
— 4K: 35-50 Mbps for sharp output, 25-30 Mbps for smaller files (still good)
— Codec: H.264 (best compatibility), H.265/HEVC (50% smaller files at same quality but limited platform support)
Cause check: if source video is itself low-resolution (e.g., Zoom recording at 720p), no export setting upscales it cleanly. The 'blurry' is baked in.
Fix for upscaling: AI upscalers (Topaz Video AI, ON1 Resize) before re-importing. Or accept the source quality.
Also check: export resolution matches project resolution. Don't export 1080p from a 720p project — Descript stretches and softens.
Step 4
One speaker too loud, another too quiet. Apply normalization per track. Or use Audio Match → Match to Reference Track.
Multi-speaker projects with different mics + setups produce uneven levels.
Manual fix per track: select each speaker's track → Audio panel → Gain → adjust. Quiet speaker +6 to +10 dB; loud speaker -3 to -6 dB.
Auto fix: Tools → Audio Match (or Audio Levels Match depending on UI version). Pick a reference track (the well-leveled one). Descript matches other tracks to it.
Loudness normalization (per track): set each track to peak around -3 dB and average around -16 LUFS. Use the Loudness meter in the Audio panel.
Master loudness for podcast: -16 LUFS total. For YouTube: -14 LUFS. For broadcast: -23 LUFS. Match your destination's spec.
Compression: light compression on loud speakers evens out their peaks. Descript's compressor is in Effects panel — start with default settings.
Step 5
Over-applied Studio Sound. Reduce intensity. Remove EQ stacking. Don't apply Studio Sound to music tracks.
Robotic / compressed audio usually means Studio Sound was applied too aggressively or stacked with other effects.
Fix 1: open the track → Effects → Studio Sound → reduce intensity to 50-70% (default is 100%). Compare before/after.
Fix 2: remove stacked effects. If you applied Studio Sound + Compression + EQ on the same track, the compounding produces over-processed audio.
Fix 3: don't apply Studio Sound to music tracks. Algorithm assumes voice; on music it produces weird artifacts.
Fix 4: source recording is the issue (recorded too hot, clipping). No post-processing fully recovers clipped audio. Re-record if possible.
Fix 5: try Descript's Voice Isolation instead of Studio Sound for noisy environments. Sometimes a gentler tool produces cleaner results.
Step 6
Stuck export: cancel, restart Descript, retry. Failing export: lower resolution/bitrate, free up disk space, check for corrupted clips.
Export stuck at X% for >10 minutes:
— Cancel the export (Publish panel → Cancel).
— Quit and restart Descript Desktop. Sometimes the render queue gets stuck.
— Retry the export.
If still stuck:
— Lower the export resolution. 4K exports occasionally fail on lower-end machines; 1080p succeeds.
— Check available disk space. Render needs 2-3x the final file size as temp space. Free up disk if low.
— Check for corrupted clips in the project. Symptom: export consistently fails at the same timestamp. Trim around that clip and retry.
If export completes but file won't play: try a different format (H.264 vs H.265 toggle), or open in a different player.
Persistent export failures: contact Descript support with the project ID and export log. They can sometimes diagnose server-side issues.
Step 7
Match codec to destination. YouTube: H.264. Apple Podcasts: AAC for audio. Some platforms reject H.265. Test before scheduling at scale.
Some exported files play locally but fail on the destination platform.
Symptom 1: YouTube rejects upload or processes forever. Fix: use H.264 video codec, AAC audio codec. Avoid H.265 if platform doesn't support.
Symptom 2: podcast host rejects audio. Fix: export MP3 at 128 kbps stereo (Apple Podcasts spec) or 192 kbps for higher quality. Avoid VBR; use CBR.
Symptom 3: file plays in QuickTime but not in browsers. Fix: ensure 'Web optimized' or 'Fast start' is enabled in export options. Moves metadata to file head for streaming.
Symptom 4: file too large for upload. Fix: lower bitrate (8 Mbps instead of 12 for 1080p) or use H.265 if destination supports.
Always test with one upload before bulk-scheduling: post one episode/clip to each destination, verify it plays correctly, then schedule the rest.
Common mistakes
Mixed frame rates in source files
What goes wrong: A/V sync drift accumulates over the video. By minute 30, audio is 1-2 seconds ahead of mouth movements. Unplayable.
How to avoid: Standardize source files to one frame rate (30 fps recommended). Convert in Premiere/Final Cut/FFmpeg before importing if needed.
Over-applied Studio Sound
What goes wrong: Audio sounds artificial and over-compressed. Listeners detect 'AI processing' even when they can't articulate what's wrong. Trust drops.
How to avoid: Reduce Studio Sound intensity to 50-70%. Apply only to voice tracks that need it. Compare before/after every time.
Bitrate too low for export
What goes wrong: Video looks soft despite sharp source. YouTube re-compresses on top, making it worse. Looks unprofessional.
How to avoid: 1080p: 12-15 Mbps. 4K: 35-50 Mbps. Check Advanced export settings — defaults are sometimes conservative.
No level matching between speakers
What goes wrong: Host's mic is loud, guest's is quiet. Listener constantly adjusting volume. Bad experience, listeners drop off.
How to avoid: Tools → Audio Match → Match to reference track. Or manually adjust per-track gain. Target -16 LUFS for both.
Exporting to a too-new codec for the destination
What goes wrong: Upload to YouTube takes 4 hours to process and shows in 480p for the first day. Looks broken to subscribers checking the launch.
How to avoid: H.264 for YouTube, MP3/AAC for podcasts. H.265 only if you control the playback environment.
No test upload before bulk-publishing
What goes wrong: All 12 weeks of scheduled content fails on platform because of one codec issue. Manual re-export and re-upload required for everything.
How to avoid: Always test one upload to each destination. Verify it plays. Then bulk-schedule. 5 min upfront saves hours of re-work.
Recap
Done — what's next
How to produce a podcast end-to-end in Descript
Read the next tutorial
Hand it off
Diagnosing export issues is one-time work. Owning ongoing export quality across weekly content production — managing source standards, export settings, destination testing — is a job. A vetted video editor on EverestX runs this from $14-16/hr — typically $400-900/mo as part of broader production work.
See video editor rates
Three usual causes: (1) export bitrate too low — bump to 12-15 Mbps for 1080p or 35-50 Mbps for 4K, (2) export resolution mismatched with project resolution — Descript stretches/softens when upscaling, (3) wrong codec for destination — YouTube re-compresses H.265 aggressively; use H.264 for upload.
Tools → Audio Match → choose the better-leveled track as reference. Descript adjusts other tracks to match. Verify each track peaks around -3 dB and averages -16 LUFS. Apply light compression if peaks still vary widely after matching.
Podcasts (Apple Podcasts, Spotify): -16 LUFS integrated. YouTube: -14 LUFS integrated. Broadcast/TV: -23 LUFS. Set master loudness target in Descript's loudness meter before export. Platforms re-normalize to their target if you're off; you sound thin or loud relative to others.
Usually one of: (1) disk space ran out during the final write step — free up disk and retry, (2) a corrupted clip near the end of the project — trim around it and retry, (3) Descript's render service hung — quit and restart Descript Desktop. Persistent: contact support with the project ID.
Not in a single export. Workaround: export Section A at one setting, Section B at another, then concatenate in Premiere or FFmpeg. Rarely necessary — usually one quality setting works for the whole project. The exception: 4K hero opens with 1080p body content (export 1080p for the body, upscale-merge the 4K open).
Descript
Most podcasters bounce between 4 tools: Riverside for recording, Descript for editing, Audacity for cleanup, Canva for show art. Descript can do 80% of it in one app. Here's the full episode workflow.
Descript
Editing is half the job. Publishing, sharing review links, gathering feedback, and shipping cleanly is the other half. Descript has solid publishing + collaboration features — here's how to use them so projects don't get stuck in 'waiting for feedback' purgatory.
Descript
Descript's killer feature: edit your video by editing the transcript. Delete a word in the doc, the audio/video deletes too. This walks the full workflow — Filler Word Removal, Word Find & Replace, Strikethrough, and Studio Sound — that makes Descript 5-10x faster than traditional editors.
Descript
Most content creators hit the DIY Descript ceiling at 10-20 episodes. Editing eats 6-12 hours/week per episode. Quality plateaus. Cadence slips. Here's the honest framework for when to bring in a podcast or video editor.
Loom
If your Looms have echo, lag, blurry video, or audio drift, the cause is almost always one of 7 specific issues. This walks each one with the exact fix — no more 'reinstall Loom' generic advice.