AI multicamera podcast editing: from synced sources to a story

01

Prepare one recording as one source group

Gather the camera files and any separate audio recording for the same session. Use clear filenames so you can recognise the wide shot, host and guest angles. Check that the files cover the exchange you intend to use and that the audio is intelligible before starting ingest. A missing answer cannot be recovered by a better brief.

In Anjin Media, a source group supports up to eight total files, including audio. Eight is a file limit, not a promise of eight cameras plus extra audio. The upload documentation explains camera and audio roles. Each camera’s footage consumes source minutes: two cameras recording for 60 minutes represent 120 source minutes.

For two or more camera files, wait until the group is ready and synchronisation has completed. The archive reference exposes sync status separately from transcription. If sync is incomplete, investigate the source group before asking for a cut. Once a render is available, check visible speech against the audio at the beginning and later in the selected exchange.

02

Find a complete exchange, not just a strong sentence

Choose the audience and the question this edit should answer. Listen before and after the candidate passage. Does the answer rely on a question you planned to remove? Does “that approach” refer to an earlier example? Does a later caveat change what the opening sentence means?

A useful short story might contain a problem, a decision and a consequence. It may need a quiet explanatory sentence more than an emphatic sound bite. Keep the exchange narrow enough to resolve one point within the requested duration. For more on that choice, see the podcast editing overview.

03

Brief the story and the camera decisions

State what must be included, what can be removed and which context must survive. Ask for a reviewable plan before committing to the render. Treat camera directions as editorial intent to inspect, not a guarantee that the source contains suitable coverage.

The multicamera podcast edit brief is a reusable starting point. If the goal is a longer first assembly for an editor, use the multicam rough-cut workflow and leave room for a finishing pass.

04

Review meaning before polishing the switches

Read the cut plan in order, checking the source timecodes, transcript and chosen camera for each segment. Confirm that the question and answer still belong together. A reordered passage must not turn a hypothetical comment into a statement of fact or remove the condition on which advice depends.

Then inspect the angle choices. Stay with an explanation long enough to follow it. Switch when a new speaker, a useful two-person view or a supported transition helps comprehension. A reaction shot should come from the relevant moment; an unrelated nod can falsely suggest agreement. The multi-angle editing feature describes the capability, while shot-aware continuity covers why boundaries matter.

05

Watch the render as a new listener

Review the whole output with headphones, then on the intended viewing device. Check lip sync, audible joins, interrupted words, abrupt camera changes and captions. For a vertical version, verify that the current speaker remains in frame and that captions do not obscure their face. A correct plan still needs this visual and audio pass.

Request one precise revision at a time: restore the question, retain the qualification, shorten the introduction or change an awkward transition. Recheck the revised output rather than assuming a local change left the story intact.

• All camera and audio files belong to the same recording; the group is ready.
• The opening identifies the subject without requiring the full episode.
• The answer resolves the question and preserves important uncertainty.
• Camera changes and reaction shots match the exchange.
• Sync, audio joins, captions and framing pass a full playback review.
• The approved output retains its brief and source references.

QUESTIONS

Common questions