Sound & Finish
If the viewer strains to hear the argument, the argument has already lost authority.
- Dialogue clarity beats cinematic mud.
- Room tone is production insurance.
- Music should frame proof, not bully it.
- Export to YouTube's stated specs.
Job To Do
Make Trust Audible
Sound is where the viewer decides whether the production is competent before they can name why. Clean VO, intentional silence, readable captions, and stable exports all say the same thing: this episode was handled.
Glen Weldon's NPR guide is practical about this: collect sound deliberately, monitor with headphones, capture room tone, and do not assume the microphone hears what your brain edits out. James Alburger adds the performance layer: posture, breath, and intent change the voice before plugins touch it.
Fang's broadcast writing also belongs here. Natural sound and silence are part of the writing surface. A wall-to-wall narrator leaves no room for evidence to breathe.
Do the clarity pass before the taste pass. If dialogue, citations, captions, and source clips are not intelligible, music choice is a distraction with better branding.
Sound Priority Pyramid
Build from the bottom. If VO clarity fails, everything above it becomes expensive fog.
Capture
Fix It Before It Enters The Timeline
Record in the quietest room available. Kill HVAC, refrigerators when practical, desk rattles, chair squeaks, browser pings, and fabric noise. Then record room tone before moving anything.
Alburger's body advice is not mystical. Stand or sit with a clean breath path, mark intent before the line, and do not perform the sound of seriousness. Seriousness lives in the claim and the receipt.
When a take fails, restart from the nearest complete thought. A clean thought edit is easier than a syllable patch with different emotional weather.
Record a 20-second room tone strip for every VO session and label it with the date, mic, room, and project. Put it in 01_sources/audio, not in the editor's memory.
Export-setting threads often argue that YouTube will re-encode everything, so the safest move is to upload a clean, high-quality source within practical bandwidth. Treat that as user-experience signal. The actual spec authority remains YouTube's own recommended upload settings.
Marked addition: forum experience, not platform documentation.
Mix
Use Sound Design As Argument Control
The Blindspot sound bed should clarify hierarchy: VO first, source audio second when it carries proof, music third, texture last. If the score makes the viewer feel more than the evidence supports, lower it or delete it.
Use silence deliberately. A pause after a document reveal can do more than a swell. It tells the viewer the receipt is allowed to land without theatrical hand-holding.
Keep foley and texture literal. Typewriter, paper, projector, and mechanical sounds can help the dossier feel, but they should not turn every source into a spy trailer. The goal is precision, not cosplay.
VO tells the path. Source audio proves the world. Music manages pressure. Texture stays in the back seat.
Captions
Captions Are Editorial, Not Clerical
Auto-captions are a draft. They are not a finish. The damage is highest where The Blindspot is most likely to need precision: proper nouns, acronyms, foreign names, dates, agencies, and legal terms.
Build a term sheet for recurring names and concepts. Use it across SRTs, lower thirds, title cards, and descriptions. Inconsistent spelling is not just ugly. It makes the channel look less source-safe.
Caption source audio with speaker and context when it matters. If the viewer joins with sound low, the caption should still show who is speaking and why the quote matters.
The upload is a private proof for internal timing only. Anything public, unlisted for review, or sent to collaborators gets caption and name checks.
Export
Finish To The Platform, Keep A Master
Anderson's workflow keeps export and archive as distinct steps. That matters. Keep a high-quality master for the archive, then make the platform upload file from that master or from the locked timeline.
YouTube's recommended upload settings are the platform source of record: MP4 container, H.264 video, progressive scan, same frame rate as recorded, 48 kHz audio sample rate, and BT.709 for SDR uploads. The recommended SDR video bitrate is 35-45 Mbps for 4K at standard frame rates and 8 Mbps for 1080p at standard frame rates; stereo audio is listed at 384 kbps.
| Export Check | YouTube-Safe Default | Source |
|---|---|---|
| Container | MP4, fast-start behavior where available. | YouTube Help |
| Video | H.264, progressive, High Profile, 4:2:0. | YouTube Help |
| Frame rate | Upload at the same frame rate recorded. | YouTube Help |
| Audio | AAC-LC or supported equivalent, 48 kHz sample rate. | YouTube Help |
| Color | BT.709 for SDR uploads. | YouTube Help |
| Archive | Keep a high-quality master plus project files and SRT. | Anderson / Binder |
Do not treat YouTube's encoding page as a loudness spec. Use it for container, codec, color, frame rate, and bitrate. Loudness targets belong in the local QC workflow and should be tested against real playback.
"Sound finish is not polish. It is proof that the episode can survive playback."
The Blindspot Almanac working rulePrint View
Just The Checklist
Sound & Finish Checklist
- Kill room noise before recording.
- Record and label room tone for every VO session.
- Restart failed takes from a complete thought.
- Check VO intelligibility before music taste.
- Keep source audio audible when it carries proof.
- Use silence after major receipts or consequences.
- Hand-check captions for names, dates, agencies, and acronyms.
- Export a high-quality master and a platform upload file.
- Match YouTube's current upload recommendations for codec, frame rate, audio sample rate, and color.
- Archive master, SRT, source notes, project files, and upload file in the episode folder.