Auto captions
Speech on the timeline becomes timed captions. A second voice gets its own speaker label. Change a word or a line break, then burn the titles in or write SRT and VTT next to the master.

AI tools
Captions, voice, transcript edits and cleanup sit on the same timeline as the grade. You approve every change. Nothing is baked until you export.
Speech on the timeline becomes timed captions. A second voice gets its own speaker label. Change a word or a line break, then burn the titles in or write SRT and VTT next to the master.
Read the interview as text. Delete a false start, an um, or a repeated line and the picture closes up. Put a line back if the cut was wrong. A word in the transcript jumps the playhead.
Type the line, choose a voice, and place it under the picture. Pace, pitch and pauses sit on the clip. Music ducks while the line plays, in the same mixer as every other track.
Record or import a short clean reference. Later pickups follow that voice, for the line the talent could not stay to re-record. Creator and Pro.
Take the transcript into another language and lay a read that hits the same pictures. Captions in the new language stay locked to the dubbed track. Creator and Pro.
Dead air, ums and false starts are marked on the timeline. Accept or reject each one. Nothing is removed until you confirm the cut.
Pull a person or a product off the shot. The matte is a normal clip: garbage-matte the edges, grade it, and composite it on the same timeline.
Reduce room tone and steady a noisy line. A lighter pass is on every edition. Finishing-grade dialogue enhance, upscale and gaze correction (when someone is reading a prompt) are on Pro. The camera original stays in the bin.
Describe the cue and get a music clip cut to the sequence length: tension under a minute, a light bed under a product line. It is a temp. Swap it for licensed music before delivery.
The music track suggests cut points. Keep the ones that fit the scene and ignore the rest. They are markers, not a montage you cannot undo.
On Pro, point a script at a bin. Night Sterling strings a first assembly from matching lines and clip names, with voice and captions already laid. You still decide the cut.
Paste a script on any edition. Night Sterling lays the voiceover and captions on the timeline so you can drop picture under them. The rough cut from a large bin is the Pro step above.
Captions, transcript edit, core voices, script to voice and titles, silence removal, background removal, temp beds, beat markers, 1080p.
Everything in Free, plus translate and dub, voice match, and the longer score cues.
Everything in Creator, plus script to first assembly, dialogue enhance, upscale and gaze correction.
Vertical canvases for TikTok, Reels and Shorts use the same captions and voice.