Captions, word by word
Eight styles, from bold yellow highlights to a karaoke plate. Each word lights up as it is spoken. The preview is drawn by the same code as the exported file, so what you see is what you get.
Drop in a talk, a podcast or a lesson. TenunKlip writes animated captions word by word, finds the clips worth posting, and keeps the speaker in frame when it crops to vertical.
Windows, macOS and Linux · No account, no watermark, no upload · MIT or Apache-2.0
Eight styles, from bold yellow highlights to a karaoke plate. Each word lights up as it is spoken. The preview is drawn by the same code as the exported file, so what you see is what you get.
Click a word to jump to it. Click it again to type the fix. Words the recogniser was unsure of are underlined, so you know where to look.
TenunKlip finds 20 to 60 second stretches that stand on their own. A small AI on your computer scores each one for its hook, emotion and value, and says why. You choose.
For vertical and square video, the crop moves with the person talking: still while they sit still, a gentle pan when they drift, a cut when the subject changes.
Speech recognition, face tracking and clip ranking all run offline once downloaded. No account, no credits, no watermark, no monthly fee.
Speech is recognised in about 100 languages and detected automatically. Caption fonts cover languages written in the Latin alphabet, including English and Indonesian.
Version 0.1.0. Free and open source.
On first start, TenunKlip downloads a video tool and a speech model, about 250 MB. The apps are not code-signed yet: on macOS, right-click the app and choose Open the first time; on Windows, choose More info, then Run anyway.