Clip Maker

Turn a video you own into vertical 9:16 clips with burned-in captions, on your own computer.

What it does

Transcribes the speech on your computer with word-level timing. You can also use your own SRT file, or no captions.
Picks the best moments by speech density, loudness and hook-style openings. Clips never overlap.
Renders 1080x1920 H.264 clips with fit-and-blurred-background or center-crop framing, 3 caption styles and an optional watermark.
Exports an MP4, SRT and cover image for each clip, plus a ZIP with a manifest that records the source file's SHA-256 fingerprint.

No upload, no account, no API key and no cloud cost. It asks you to confirm you own the footage.

Speech recognition can mishear names and jargon, so check captions before posting.
Moments are picked from speech and loudness only; it doesn't understand the picture.
Center crop doesn't track faces.
Today it needs Python and ffmpeg installed on Windows. A one-click installer isn't built yet.

It is built and tested. Inquire for deployment details. Email us for more information.