Clipping

Podcast clips that follow whoever is talking

Two or three people at a table, one camera wide: the clipper works out who is speaking at every moment and cuts the vertical frame to them, with captions and a hook on top

Clip an episodeSee pricing

Free plan: about 15 clips a month (300 credits) · no card needed

The 9:16 crop follows the active speaker, with zoom-cuts on emphasis · Real pipeline output · footage: The Smart Agents Podcast (CC BY)

How it works

  1. 01

    Paste a link or upload

    A YouTube URL or your own file: podcasts, interviews, streams, talks, matches. Longer is fine, every minute gets scored

  2. 02

    The clipper finds the moments

    It transcribes, reads faces, voices, reactions and audio peaks, then ranks candidate clips for hook, payoff, pacing and whether they stand alone

  3. 03

    Review, tweak, post

    Each clip opens as an editable project in the Studio with captions and crop already in place. Export or schedule straight away

What you get

  • Speaker diarisation mapped to faces

    Voices are separated and matched to the faces on screen, so the crop switches on the right person, not on whoever moved last

  • Wide shots handled

    When two guests sit far apart, the frame cuts between them or uses split tiles instead of a blurry middle crop

  • Moments, not random minutes

    Laughs, strong opinions and clean standalone answers score higher than filler, so the best bits come first

  • Brand kit on every clip

    Your fonts, colours, logo and caption theme applied to the whole batch

Why podcast clips are harder than they look

A video podcast is usually two or three people at a table and a single wide camera, or a few static angles. The best line of the episode might come from the guest on the far left while the host laughs on the right, and a centre crop to 9:16 shows a microphone and two half faces. That is why most quick podcast clips look cheap, and why they underperform

The podcast clip generator solves the framing first. It tracks each face across the whole episode, separates the voices with speaker diarization and maps each voice to a face, so the vertical frame cuts to the person who is actually speaking. When both people matter, it uses split tiles rather than a blurry middle crop

Finding the moments worth posting

Long conversations have a lot of setup and a few moments of payoff. The generator reads the full transcript and scores overlapping windows for hook, payoff and whether the clip stands alone without the previous ten minutes. Strong opinions, clean answers, stories with an ending and moments with laughter score higher than introductions and housekeeping

You choose how many clips and how long. The strongest candidates come first, each with the reason it was picked; clips the critic rejects are only used when they are needed to reach your minimum

Turning one episode into a week of Shorts

Each clip opens as an editable project with captions, a hook line and the speaker-following crop already in place. Apply your brand kit to the whole batch, fix any word in the captions, then export the clips as MP4 for YouTube Shorts, Instagram Reels, TikTok and LinkedIn

For audio-only shows, there are no faces to track; use the music and captions flow to turn a section into a captioned clip with a motion background

Compare AI clippers

FAQ

Questions people ask

Updated October 2026

How do I turn a podcast into Shorts?
Upload the episode or paste its YouTube link. The podcast clipper transcribes it, separates the speakers, scores every moment for hook and payoff, and cuts the best ones into 9:16 clips that follow whoever is talking, with animated captions. Review them, tweak any clip in the Studio, then export or schedule to Shorts, Reels and TikTok
Does it work with a single wide camera?
Yes. That is the case it is built for. It tracks each face across the episode, matches voices to faces, and cuts a vertical frame to the active speaker, falling back to split tiles when two people need to be on screen together
Can I clip audio-only podcasts?
Audio-only episodes can be turned into captioned clips with a waveform or motion-graphics background from the music and captions flow, since there are no faces to track
What does it cost?
Clip extraction is paid in credits (about 20 per clip), estimated before you start. The free plan includes 300 credits a month, about 15 clips; your first 3 clip exports are free, then 10 credits each. The $10 Clipper plan gives about 50 clips a month at 1080p with no watermark and 300 clip exports a month

Guides

More tools