Use case · Archive to content
Index once.
Create repeatedly.
Most organisations sit on years of recorded sessions and still make every new video from scratch. Load each session into Anjin Media once - transcribed to the word, speakers separated - and it stays addressable. To turn a video archive into content after that, you write a brief, name the sessions it should draw on, and read the plan that comes back.
Up to four sessions can feed one brief, and a session can feed as many briefs as your plan's compositions cover.
One video or an entire archive.
Archive to content product demonstration
The problem
Recorded, then locked away.
The keynote was shot. The panel was shot. The all-hands is shot every month. Then the files go wherever files go, and six months later a team that needs ninety seconds of the founder on culture starts by asking who might still have the drive. The material exists; the cost of reaching into it is what kills the idea.
Search alone leaves the gap open. Knowing the sentence exists still leaves someone to pull the file, find the moment, cut it and build something publishable around it - the distance between found and finished is where archive projects stall.
Packages, provenance and security for large archives are covered in depth on the enterprise page →
What you make
The edits an archive owes you.
Creating edits from a media library stops being a project and becomes a habit: each of these is one written brief against sessions already indexed.
The recruiting film
The founder on culture across three years of sessions, cut to ninety seconds for the careers page - sessions named in the brief, the running order argued by the plan and read back before anything renders.
1:30 · 16:9 · THREE SESSIONS
The proof reel
Customers making the case in their own words, drawn from the panel session for the sales deck. No script, no re-record: the evidence already exists on tape.
0:40 · 1:1 · DIARISED SPEAKERS
The weekly drumbeat
One flagship session, briefed again each week - a different claim, a different audience, the same indexed source. The tenth edit needs no new footage, only a new paragraph.
ONE SESSION · BRIEF AFTER BRIEF
The workflow
A session becomes addressable.
A session loads as a source group: up to four camera files, each up to 16 GB and four hours, treated as one recording. Ingest transcribes every word with timecodes and separates the voices; once the group reads ready, any sentence in it can be reached from a brief.
That work happens once per session. Every brief afterwards is written against the library and names the sessions it draws on - like the recruiting-film brief here, which names three - and returns a plan whose segments keep their source timecodes. The archive's paper trail survives into everything cut from it.
Loading is the only heavy step, and it runs in parallel: groups ingest independently and a webhook announces each one the moment it turns ready. The tenth brief, like the first, starts at a keyboard.
“Ninety seconds for the careers page: the founder on what we hire for, in her own words across the years. Open on the 2024 keynote, close on the answer she gives now.”
Where it fits
Scoped for teams, honest about scope.
CHOOSE IT WHEN
- Your sessions are spoken and worth reusing: town halls, keynotes, panels, interviews, webinars.
- Several teams need finished edits from the same material, on their own schedules.
- Every edit must show its working - segments checkable against sources, an EDL for the archivist.
- The archive should be callable from your own systems, by people and by agents.
LOOK ELSEWHERE WHEN
- You need undirected discovery - “surface the best moments across everything we own”. A brief here draws on the sessions it names, up to four at a time.
- The footage is mostly wordless; retrieval runs on the transcript.
- Rights and clearance tracking must live inside the tool. Rights metadata is on the roadmap; today clearance stays with your records.
SCOPE, PLAINLY
A brief composes from up to four named sessions, so an archive-wide sweep is a claim you will never read here. Access is account-wide today - per-library permissions are on the roadmap - and planning is asynchronous: briefs are acknowledged at once and plans follow when composed. Source minutes are counted once, as you bring footage in. Briefing, plans, revisions and renders use none.
The whole loop is callable - a REST API and an MCP server - so an intranet form or an agent can be the archive’s front door. Newsroom-scale teams have a page of their own, and plans and packages are printed in full on pricing.
The objection
The question every buyer asks.
- Can one prompt search our whole archive at once?
- No. A brief here draws on the sessions it names - up to four at a time - and that scoping is deliberate: what comes back is checkable against exactly the sessions it drew on. The library's job is to make naming them quick: each session shows its date and duration in the list, and opens to its named speakers, talk time and full transcript. Pointing a brief at the right material takes a minute, and what a brief cannot do is quietly pull from somewhere you did not intend.
Questions
- How do I turn a video archive into content?
- Index each recorded session once as a source group - it is transcribed with word-level timecodes and its speakers are separated. From then on, write briefs that name up to four sessions each, set a length between 10 and 600 seconds, and review the plan that comes back before rendering. Each render arrives with its captions (SRT), a machine-readable edit decision list and a QC report.
- Can I create edits from a media library without re-uploading footage?
- Yes - that is what indexing buys. A session stays addressable after the project that loaded it ends, so the fifth brief against a keynote takes the same work as the first: writing it. Nothing is re-uploaded, re-transcribed or re-watched.
- How much video can be indexed?
- Self-serve plans meter source minutes as footage comes in - from 120 minutes on the 14-day trial to 1,500 a month (25 single-camera hours) on Studio. Enterprise packages start at 2,000 archive hours, with Network sized to the organisation. Each file caps at 16 GB and four hours, with up to four cameras to a session.
Put your archive to work.
A demo runs on your own material: load one session, brief one edit, read the plan it returns.
Index once. Create repeatedly.