How to cut several vertical highlight clips with subtitles
Give Vizard Agent a long video and say how many vertical clips you want. Vizard Agent transcribes it, finds the strongest moments, tests the 9:16 crop on real frames to confirm the subject is not cut off, and burns in subtitles on clips that fill the frame edge to edge with no black bars.
What do you type into Vizard Agent?
Say the count, the shape and the subtitle language in one line. Vizard Agent chooses the moments from what is actually said and shown, so you don't need timestamps unless there's a section you want covered. Say no black bars explicitly: it means filling the frame rather than shrinking the wide picture into it.
Prompt
What does Vizard Agent actually do?
It doesn't assume the crop is safe. A centre crop that beheads the speaker looks fine in a progress log, and a missing font draws empty boxes instead of throwing an error, so Vizard Agent renders a test frame at the target shape and finds the font before any clip is committed. In the session behind this article it:
- Probed the video's streams and duration, transcribed the audio with timestamps, and looked at frames from several points to understand the composition.
- Built a shot map of the video to find precise cut points.
- Rendered test crops at 9:16 and inspected them to confirm the subject stayed in frame.
- Wrote a script to slice the word-level transcript and shift each clip's timestamps to start at zero, then generated the subtitle files for all five clips in parallel.
- Listed the installed font families and found the exact font file path so the burn-in rendered correctly. One delivered clip probed at 540x960, 25fps H.264 with AAC audio, 17.0 seconds long.
When does this not work well?
Reframing wide footage to vertical throws away most of the picture. Vizard Agent follows the subject rather than the centre, but some compositions don't survive. Two people side by side won't both fit, scoreboards and lower thirds at the edges go first, and a low-resolution source sets a low ceiling. A moment that needs ten minutes of context won't stand alone.
Common questions
Will the subject stay in frame? Vizard Agent tests the crop on real frames and follows the subject rather than cropping the centre blindly.
Can the subtitles be in a different language from the speech? Yes. Say which language you want and Vizard Agent generates them accordingly.
Does it work on a podcast recording? Yes. Two-person framing is the case to watch, so tell Vizard Agent which speaker to follow.