Skip to main content
[ Podcasts and interviews ]

Clips from a long episode that make sense alone

Give it a podcast episode with video. VimeoClip reads everything said, suggests moments with a clear start and finish, and rebuilds them for a vertical frame. You watch and choose.

Field reporter · Illustration
Field reporterIllustration
Podcast host at the mic · Illustration
Podcast host at the micIllustration
Talking to camera · Illustration
Talking to cameraIllustration
Illustration (stock video), not VimeoClip output · Storyblocks.

Podcasts, field reports, talking to camera

Stock footage from Storyblocks, to show the kinds of video you can bring. These are illustrations, not clips VimeoClip made.

  • Podcast host at the mic · Illustration
    Podcast host at the micIllustration

    Podcast host at the mic

    Illustration (stock video), not VimeoClip output · Storyblocks

  • Field reporter · Illustration
    Field reporterIllustration

    Field reporter

    Illustration (stock video), not VimeoClip output · Storyblocks

  • Talking to camera · Illustration
    Talking to cameraIllustration

    Talking to camera

    Illustration (stock video), not VimeoClip output · Storyblocks

What happens to an episode

Moments chosen from what was said

The system reads the transcript of the whole episode and looks for passages that stand alone. Cuts favour the end of a sentence, using punctuation and pauses in the speech.

  • A job with no length set uses about 40–75 seconds for a podcast and about 35–70 seconds for an interview.
  • The Podcast preset in the dashboard uses 60–89 seconds, and you can change it before running.
  • A clip whose cut could not be anchored to the transcript shows a warning on its card.

Two people in one vertical frame

The Split layout stacks two speakers as top and bottom halves. Captions sit on the seam between the halves, where they cover nobody.

  • Both faces must appear in the same frame in at least half of the sampled frames.
  • Shot and reverse shot does not use Split, because stacking would show one person twice.
  • When the cut goes to one person, the frame returns to following that face.

Cuts that follow the speaker

The Speaker Cut layout switches the frame to whoever is talking. The two-person interview preset in the dashboard uses it, with one-line captions and a headline for the first three seconds.

Ask for the moment you remember

If you remember a guest telling a particular story, type a description before running. The AI searches the spoken words, so describe what was actually said.

What podcasts tend to add

Captions

Word-by-word captions with your choice of look and keyword highlight colour.

See captions

Episode summary

Gather the main points into one video, in the order they came up.

See summary

Dubbing

Take a clip into 31 languages for other audiences. Needs an ElevenLabs key.

See dubbing

Good to know first

  • The dashboard upload box takes video only. An audio-only episode has to be exported as video first.
  • Automatic layout sends 12 sample frames of the video to the AI model, so it needs a model that can read images. Without one, you choose the layout yourself.
  • Split and Speaker Cut depend on face detection. A distant camera or covered faces can fall back to the plain frame.
  • Dubbing needs an ElevenLabs key and is enabled by an admin.
  • Each job uses credits based on the length of the source video. A new account can start with 0 credits.
  • The footage on this page is stock video from Storyblocks, shown as an illustration, not VimeoClip output.

Try it on your latest episode

Paste the episode link to start.