Checklist the per-video bar — engine/sim
83.4/100
quantitative quality · weights learn from your reviews (
engine.sim.memory review thread-build-log-W26-Wed good|bad)⚠ 6 flag(s) — not yet ship-ready: copy_genericlow_vo_coveragedead_airgeneric_scene · see docs/strategy/VIDEO-CHECKLIST.md
Composition comp · template family · expected output
9:16 Reelrendered1:1 Squarepending16:9 Widepending9:16 4Kpending1:1 4Kpending16:9 4KpendingGIF (SMS)pending
expected output: 1/7 rendered — same matrix the /media preview surfaces for this asset.
Composition layer × scene 6 scenes · 33.4s · comp_id + rendered still + tier + the script
| # | Layer (comp_id · still · tier) | Beat | Timecode | Motion | Logo | Audio | VO / on-screen / caption |
|---|---|---|---|---|---|---|---|
| 1 |  matches intent shared field signature-3d | open | 0–5.8s | spatial-parallax | icon·wireframe | ♪ bed_in | Week four. We shipped automated voiceover this week, a script to spoken audio in seconds. on-screen: Shipped: AI voiceover, free of a studio |
expected on screen: white ground · cone hero in the shared Signal Field · Augur leads · node-graph · spatial-parallax · icon·wireframe logo · caption bottom-left spec (the prompt): comp_id shared Signal Fieldvisual node-graphshape coneground whitetreatment wireframemotion spatial-parallaxpower summoninstrument summon→Shipped: AI voiceover, free of a studio | |||||||
| 2 |  matches intent JsonDiff template | build | 5.8–10.8s | type-on | icon·color | ♪ node_lock | It used to mean booking a booth and a voice actor for every revision. on-screen: Old way: book a booth |
expected on screen: white ground · a JsonDiff panel over a dimmed Signal Field · Augur leads · code · type-on · icon·color logo · caption bottom-left spec (the prompt): comp_id JsonDiffvisual codeshape coneground whitetreatment colormotion type-onpower summoninstrument draw-on→Old way: book a boothcurate codeLines, fileName, lines | |||||||
| 3 |  matches intent StackTrace template | diff | 10.8–16.700000000000003s | crossfade-8f | icon·color | ♪ node_lock | It broke funny. The model read our stage directions out loud, the word pause and all. on-screen: What broke: it read the stage directions |
expected on screen: white ground · a StackTrace panel over a dimmed Signal Field · Augur leads · code · crossfade-8f · icon·color logo · caption bottom-left spec (the prompt): comp_id StackTracevisual codeshape coneground whitetreatment colormotion crossfade-8fpower morphinstrument morph→What broke: it read the stage directionscurate codeLines, errMsg, errType, fix, frames | |||||||
| 4 |  matches intent BlueprintGrid template | step | 16.7–23.4s | kinetic-build | icon·color | ♪ node_lock | Now every script has a voice in minutes. Next, we lock one consistent brand voice across all of it. on-screen: Next: match voice to the brand |
expected on screen: white ground · a BlueprintGrid panel over a dimmed Signal Field · Augur leads · pipeline · kinetic-build · icon·color logo · caption bottom-left spec (the prompt): comp_id BlueprintGridvisual pipelineshape coneground whitetreatment colormotion kinetic-buildpower throwinstrument laser-trace→Next: match voice to the brandcurate rows, stages, steps | |||||||
| 5 |  first render · fix pending RankList templatedead_airgeneric_scene | proof | 23.4–28.4s | receipts-count | icon·color | ♪ node_lock | Build Log |
expected on screen: white ground · a RankList panel over a dimmed Signal Field · Augur leads · receipts · receipts-count · icon·color logo · caption bottom-left spec (the prompt): comp_id RankListvisual receiptsshape coneground whitetreatment colormotion receipts-countpower receiptsinstrument spotlight→Build Logcurate rows, statsLabels | |||||||
| 6 |  first render · fix pending shared field signature-3ddead_airgeneric_scene | resolve | 28.4–33.4s | coalescence | icon·wireframe | ♪ bed_out | Build Log |
expected on screen: white ground · cone hero in the shared Signal Field · Augur leads · coalescence · coalescence · icon·wireframe logo · caption bottom-left spec (the prompt): comp_id shared Signal Fieldvisual coalescenceshape coneground whitetreatment wireframemotion coalescencepower coalescenceinstrument coalescence→Build Log | |||||||
Format stack 3 aspects · same scenes[], re-cropped
9:16
1080×1920
Stories · TikTok · YouTube Shorts · Reels
1:1
1080×1080
LinkedIn · Facebook · Instagram
16:9
1920×1080
X/Twitter · YouTube · LinkedIn video
Channels 9 destinations
LinkedInX/TwitterYouTubeInstagramFacebookThreadsTikTokPinterestBluesky
Social captions supplemental published copy · per channel (comp_id level)
tiktokBuild log, week four. We shipped automated voiceover: a script becomes spoken audio in seconds, no booth, no voice actor per revision. It broke funny too, it read our stage directions out loud, the word pause and all. Next: one consistent brand voice. (AI-assisted)
instagramBuild log, week 4.
Shipped: automated voiceover.
Script in, spoken audio out, in seconds.
Old way: book a booth and an actor for every revision.
What broke: it read our stage directions out loud, the word pause and all.
Next: lock one brand voice.
#buildinpublic #aivoice #contentops #voiceover #shipit
linkedinBuild log, week four.
Shipped: automated voiceover. A script becomes spoken audio in seconds. That used to mean booking a booth and a voice actor for every single revision.
What broke, honestly: the model read our stage directions out loud, the word pause and all. We laughed, then fixed the input.
Next up: lock one consistent brand voice across everything we publish. Progress over polish, in public.
xBuild log wk4: shipped automated voiceover, script to audio in seconds. No booth, no actor per revision. It broke funny, read our stage directions out loud, the word pause and all. Next: one brand voice.
facebookBuild log, week four. This week we shipped automated voiceover, turning a script into spoken audio in seconds. It used to mean booking a booth and a voice actor for every revision. It broke in a funny way too: the model read our stage directions out loud, including the word pause. Next, we lock one consistent brand voice across everything. Follow along.
threadsBuild log, week four.
Shipped: automated voiceover. Script in, spoken audio out, in seconds. No booth, no actor per revision.
What broke: it read our stage directions out loud, the word pause and all.
Next: lock one consistent brand voice across all of it.
pinterestBuild log week four: how we shipped automated AI voiceover that turns a script into spoken audio in seconds, replacing studio booths and per-revision voice actors. Includes the funny bug where the model read our stage directions out loud. A behind-the-scenes content automation diary.
blueskyBuild log wk4: shipped automated voiceover, script to audio in seconds. No booth, no actor per revision. It broke funny, read our stage directions out loud, the word pause and all. Next: one brand voice.
youtubeTitle: Build Log Week 4: We Shipped Automated Voiceover (And It Broke Funny)
This week we shipped automated voiceover that turns a script into spoken audio in seconds. It used to mean booking a booth and a voice actor for every revision. It also broke in a funny way: the model read our stage directions out loud, the word pause and all. Now every script gets a voice in minutes. Next, we lock one consistent brand voice across everything we publish.