Skip to content

A clip produced by ClipFlap, untouched

The clip below is a real podcast clip, straight out of the pipeline: moment picked by the AI, tracked framing, silences trimmed, word-by-word captions burned in. Nothing was retouched by hand.

With the original sound: listen and check the captions against the voice.

Demo: a podcast clip cut and captioned automatically

Length
1:07
What ClipFlap did
A full episode analysed, one peak kept, the framing follows the speaker, silences are trimmed and every word is aligned to the voice. 1080×1920 output, ready for TikTok, Reels and Shorts.
Try it on your video10 free minutes, no card.

How this clip was produced

  1. Selection: the whole episode is transcribed word by word, then the model keeps the passages that stand alone, with a clean start and end.

  2. Framing: the speaking face is tracked shot by shot and the clip is reframed to 9:16; with two voices, the screen is split.

  3. Captions: every word is aligned to the voice and burned in with the typography of the language. The clip is produced as is, untouched.