Why Aperture
Aperture exists because editorial time is wasted in the gaps: scrubbing for moments, round-tripping feedback, copying files between tools, and waiting for renders. We own the whole loop — timeline, transcript, library, review, render — so the gaps are gone. This page is the honest argument for why that makes teams faster than anything else on the market.
The old way: scrub proxies and hope. Aperture: moment search over every transcript in the library, ranked by relevance, one click from result to timeline. Finding becomes reading.
The old way: hunt for the blade, ride the playhead, chase frames. Aperture: the transcript is the edit — select the words you don't want, press cut. Silence and filler removal is one operation, not an afternoon.
The old way: exports, file transfers, feedback in email, confusion about versions. Aperture: one guest link — expiry, password, watermark — comments pinned to the frame, threads resolved, sign-off recorded in the audit trail.
The old way: re-edit the same content per platform. Aperture: LLM-ranked highlights, a 9:16 reframe preset, styled captions, auto-duck music. One master, many winners.
The old way: renders block the machine; the workstation is the bottleneck. Aperture: the server renders while you keep editing — from anywhere, on anything.
The design decisions
Most tools bolt a transcript onto a player. We own both ends: text operations are real timeline ops (razor, ripple) executed by the same engine that plays and renders. That's why "cut this sentence" is one op — not a transcript suggestion you re-do by hand.
The client composites live with WebCodecs; the server conforms with ffmpeg. You get instant feedback and consistent, professional exports — with zero installs on any device, anywhere.
Collections, smart (saved-search) collections, tags, sha256 duplicate detection, and transcripts are queryable. The library finds things for you instead of hiding them from you.
We build AI where it compounds with the timeline — text editing, search, captions, highlights, auto-tagging — and we generate media where it sells: images and video on Pro. We don't pretend a chat window is an editor.
OTIO and project JSON export on every plan. Realtime co-editing with presence. API tokens and webhooks for machine integration. If you leave, your work leaves with you.
Audit trail with sign-off export, per-collection permissions, internal-only notes, SSO via OIDC, metering. The governance a company needs is not an enterprise upsell bolted on later.
Review tools don't edit. Web editors don't do governance or archive. Transcript editors don't run a render farm. Aperture is the whole loop — ingest, search, cut, review, deliver, automate — in one browser tab.
| Capability | Aperture | Review tools (Frame.io-class) | Web editors (CapCut-class) | Transcript editors (Descript-class) |
|---|---|---|---|---|
| Real multi-lane NLE | ✓ | ✗ | Basic | Single-track feel |
| MAM + review + NLE + API in one product | ✓ one loop | Review only — can't cut | Weak MAM | No archive API |
| Search result opens the editor at that second | ✓ deep-link to timecode | Transcript search only | ✗ | Own docs only |
| Preview = render (same edit ops client & server) | ✓ | ✗ | Approximate | Approximate |
| Karaoke captions with word timings, burned at export | ✓ ASS/libass | Basic | Web-only trick | ✓ |
| Realtime multi-user timeline with presence | ✓ in a tab | Comments only | Co-edit, no timeline | ✗ |
| Per-lane audio stems at export | ✓ every lane | ✗ | ✗ | Partial |
| Interchange: EDL + FCPXML + OTIO | ✓ all three | ✗ | ✗ | ✗ |
| −14 LUFS loudness-normalized delivery, one click | ✓ | ✗ | Approximate | Partial |
| Watch folders + resumable ingest, self-hosted | ✓ your server | ✓ SaaS only | ✗ | ✗ |
| Rules engine + signed webhooks + audit export | ✓ via API today | Enterprise tiers | ✗ | Partial |
| Guest review where clients decide: sketches on the frame, approval states, internal-only notes | ✓ | Comments, no sketch | ✗ | Partial |
| Recordings arrive as layers — screen + camera as separate tracks, re-layoutable | ✓ in-browser | ✗ (Loom: flat file) | ✗ | ✗ |
| API-first with published OpenAPI docs | ✓ /docs, 33 paths | Enterprise tiers | Partial | Partial |
| One tool, one price | ✓ | ✗ | ✓ | ✓ |
"We are the cheapest way to get Descript's text loop and Frame.io's review loop in one place — because we own both the timeline and the transcript engine."
— Aperture product thesisFree plan, open registration. Your first sequence is one click away.
