Edit a multicam podcast automatically in Premiere Pro.

A video podcast with two or three cameras means hours of footage per angle to sync, then switch by hand at every turn in the conversation. On a 90-minute episode, lining up the claps and cutting camera by camera can take longer than the rest of the edit.

It's exactly the kind of mechanical task automation does well: spot who's talking and cut to the right angle, without fatigue or hesitation.

Many tools switch cameras based on loudness: the loudest track wins. It works, but it struggles when two people talk at once, when one person's mic picks up the other's voice, or when someone reacts loudly without taking the floor.

Tools that try to identify the speaker rather than the audio peak do better on fast back-and-forth. It's the first thing to check when you compare AutoPod, AutoCut Podcast, FireCut and Rushly on this point. Rushly reads each mic separately, accounts for one person's voice bleeding into the other's mic, and enforces a minimum shot length so it doesn't cut on every “yeah”.

You start from a sequence where cameras and mics are synced. The Multicam tab detects who's talking on each mic and shows you the cut plan before creating anything, with three pacing presets (Calm, Balanced, Dynamic). It then creates a new sequence: cameras stacked, off-air shots disabled, reframes kept. Your original sequence stays untouched.

Everything runs on your machine: no upload, no credits used. On a real 2 h 11 podcast (two people, two isolated mics), the sequence came out in about 8 seconds on the first analysis, 1 to 3 seconds on later runs, and the right person was in close-up about 98% of the time by our measurement.

The limits, stated plainly: you need one mic per person (two mics on a left/right stereo track is handled; a single mono mic for two people isn't). Rushly stays on the person talking and does not do reaction shots of the listener. And the result is a sequence with stacked cameras, not a native Premiere multi-camera sequence.

Whatever the tool, automatic multicam gives its best result with one mic per person (separate audio tracks). That's what lets it know exactly who's talking at every moment. With a single shared track, detection gets approximate, or impossible.

Once your angles are cut, the longest part is left: trimming the episode down. In Rushly, that's the next step: AI Cleanup, the semantic rough cut (silences, hesitations, bad takes), right inside Premiere. Rushly was built and tuned on French speech first. It also understands English, Spanish and Portuguese: test it on your own footage with the Free plan before you commit.