Create AI Covers in One Click

What Ribbi did

The user gives one song link and one sentence. Ribbi downloads the audio, pulls the original lyrics, recognizes the track as Stefanie Sun's "What I Miss," and rewrites the lyrics entirely in English — not line-by-line translation, but preserving the song's emotional core and structure, reorganized around Soul/R&B rhyme and rhythm. It then re-performs the song in a male Soul voice and delivers two versions — warm bass, jazz chords, a gospel-tinged texture — at roughly 255 and 246 seconds, for the user to pick the better fit.

What Ribbi did

Ribbi parses the cover version's subtitle file for the precise timing of all 59 lyric lines, then builds a complete visual narrative on top: a man alone at night, replaying a lost love, moving from doubt and hurt through deep longing, an emotional peak, self-questioning, and finally painful release. All 59 shots land line by line — each tagged with timecode, shot size, scene content and light-and-color mood, forming a directly executable storyboard.

What Ribbi did

From one word — "texture" — Ribbi derives a concrete visual direction: cinematic realism, low-key lighting, a warm yet melancholy palette, like the intimate moments in a Wong Kar-wai or Richard Linklater film — never over-dramatized, real and breathing. It then produces style reference frames for the two leads, locking character look and image texture before mass production begins.

What Ribbi did

Ribbi first delivers the 10 frames of act one, "doubt and hurt," to validate the look — a living room at midnight, a silent profile, backs turned on the bed's edge, a rainy street — light and emotional progression all aligned with the storyboard. One "keep going" later, Ribbi batches out the remaining 49 act by act: the café and starlit car roof of longing, the rain embrace and firework kiss of the peak, the empty apartment of self-questioning, the seaside sunrise and morning streets of release. 59 frames with consistent characters and a unified palette, ready to use as MV material or a mood board.

What Ribbi did

Ribbi re-pulls the subtitle timeline, assigns each of the 59 frames its duration line by line to the lyrics' rhythm, burns in the English subtitles and merges everything with the Soul cover audio into a 246-second final cut — every frame holds exactly as long as its lyric line, the imagery moving with the music from doubt to release. A complete storyboard-story MV, delivered. Want a different song or a different style? Drop Ribbi another link and go again.