Skip to content
Katechon Arc
Start creating
Menu

ANSWER

How do you make an AI music video?

Short answer

Start from a finished song, plan shots to its sections, keep the performer consistent with a reference sheet, generate clips, and cut them to the beat with lyric captions. For our High Note series we made nine original songs and cut music videos from the episodes and a few dedicated hero shots.

Song first

Everything in a music video is timed to the track, so finish the song before planning shots. For High Note, a 40-episode musical drama, we made 9 original songs with Suno and mixed them into the episodes. The music-video cuts came later, built from the episodes plus a handful of hero shots rendered specially.

Method

  1. Mark the song's sections (intro, verse, chorus, bridge) and the beats you want cuts on.
  2. Decide what each section shows: performance, story, or mood. Choruses usually return to the performer.
  3. Make a reference sheet for the performer and any recurring place. The performer's face is what viewers track.
  4. Write shots that fit the section lengths. Most of ours are 4 to 10 seconds; cut them to the beat in the edit.
  5. Render, then cut to the beat grid. Add lyric captions; they double as a hook on mute.
  6. Mix at a streaming loudness target. Our finished episodes are normalized to -14 LUFS.

Lip sync

Sung lip sync is the hard part. Video models that generate audio with the picture produce their own singing, not your recording, so a close-up of the performer will not match your vocal. Three ways around it: cut so the mouth is not the subject (wide shots, profiles, hands, crowds); use a dedicated lip-sync model for the close-ups; or design the video so the song plays over story scenes rather than a performance.

Budget

A three-minute video at about 12 seconds a shot is around 16 shots, more if you cut fast. On Seedance 2.0 Mini in Arc, an 8-second 720p clip is 76 credits, so 23 such clips are about 1,748 credits before retakes.

Where Katechon Arc fits

Arc fits music videos that tell a story with recurring characters: references keep the performer consistent and you can storyboard every section.

Arc does not sync generated mouths to your own vocal track or cut to a beat grid; use a lip-sync tool and an editor for those parts.

Related answers

How do you add music to an AI drama?Score in the edit, not in the video model: generate or license music separately, place it under the episode and mix to a consistent…How do you add voice and dialogue to AI video?Let the video model speak the lines that are on screen and add narration separately with text-to-speech.How do you make a movie trailer with AI?Lock the look and the cast first, build every location as a set of camera angles, write each shot as one action with a concrete light…How do you keep the same character across AI video clips?Give the model the same visual evidence every time: one approved reference image per character, attached to every shot that character is…