AI Slideshow Video Maker with Narration & Subtitles

What Ribbi did

Ribbi lays out multiple font sizes side by side on a 1536x2752 vertical canvas and turns the comparison into a concrete recommendation — 60-72px for longer lines, 72-90px for short emphatic beats. What was an abstract pixel number becomes a decision the user can make at a glance, locking in the bold 60px spec the final film would use.

What Ribbi did

Ribbi downloads the reference and breaks down its skeleton: 60 seconds in 9:16, 2.5D illustration in warm golden tones, 20-25 shots at 1.5-2.5 seconds each, word-pop subtitles, a deep male narrator over epic orchestral music. It then writes an original fable in the same genre — The Weaver's Tapestry, a parable of love, wealth and success woven as three threads — runs a visual-thinking pass to fix the 2.5D style, and generates 20 key frames in four parallel batches. The first narration ran 183 seconds, so Ribbi tightened the script and re-recorded it at 79 seconds to match the pacing, then assembled the film: frames timed to the subtitle track, Ken Burns micro-zoom on every shot, 0.3s cross-fades, BGM ducked to 0.25 — a 79-second vertical film in one pass.

What Ribbi did

Word-pop pacing is exactly how the reference video holds attention. Ribbi transcribes the narration segment by segment to pin down when each word lands, then consolidates that into a word-timed subtitle track — one to two words per cue, with key words like love, wealth and success getting their own beat. The text now pops in locked to the narrator's rhythm, and the recomposed final cut upgrades the subtitles from static full sentences to the dynamic word-by-word cadence that defines the format.