AI moment detection
The AI reads the transcript and listens to the audio to find openings that grab attention and thoughts that are complete on their own. Clips take their natural length, from quick 15-second highlights to stories of around 90 seconds. Every suggestion comes with a short explanation and a score, so you can quickly decide which clips to keep.
Word-by-word captions
Captions are timed word by word and can highlight key words as they are spoken. You can correct any word, change the font, size and colour, and move the captions on the screen. Transcription supports more than 30 spoken languages, including English, Finnish, German, Spanish and French, as well as automatic language detection.
Vertical reframing
Landscape recordings become 9:16 clips. In Fill mode the frame follows the speaker's face with smooth movement. Fit mode keeps the whole picture, and Split mode places supporting footage above the speaker. You can always override the framing by hand.
Eight visual styles
Each style sets caption design, highlight colours, hook titles and scene rhythm together. Switch between styles and see the preview update right away. The automatic cut option can remove pauses and add B-roll for you.
B-roll and AI images
Stock video clips can be added automatically based on what is being said. On Starter and Pro you can also generate AI images for ideas that are hard to film.
Publishing queue
- Connect YouTube and TikTok to your account.
- Add finished clips to the queue and edit the cover and title.
- Schedule posts on a calendar instead of uploading one by one.