Every episode is twenty clips you have not posted
A weekly two-hour show is roughly a hundred hours of raw material a year. Almost none of it gets clipped, because clipping is the job nobody scheduled.
For podcasts the two things that matter are handling two speakers on screen and not paying per minute of source. Soft Clipper does split-screen and speaker-tracking framing, and puts no cap on how many hours you feed it.
What actually gets in the way
Two people, one vertical frame
Split mode stacks both speakers, one above the other. Smart mode follows whoever is talking and re-picks when the camera cuts. Either beats a centre crop that shows two half-faces.
Long episodes cost the most on cloud tools
Metered services charge by minute of source video, so a two-hour episode burns an allowance whether it yields two clips or twenty. Local rendering has no such meter.
Guests and unreleased episodes
Interviews often go out under embargo. Only the audio leaves your machine for transcription — the video never does.
How the work goes
-
Drop in the episode
The recording is transcribed with timestamps, or existing captions are reused when the episode is already published.
-
Ask for the themes
Prompt search finds the segments about a specific topic — the guest's origin story, the disagreement, the practical advice.
-
Cut ten at once
Style the captions and framing once, apply to every clip, render the batch while you do something else.
Frequently asked questions
Can it handle a two-hour episode?
What about video podcasts with two cameras?
Audio-only podcast — is there any point?
Stop hand-cutting clips
Load a video, get back a set of vertical clips with captions. One payment, yours forever.