Skip to main content
Music Video listens to your song — its lyrics, mood and structure — and builds a visual story that follows it scene by scene. The vocals are transcribed for lyric captions, and an optional waveform reacts to the audio. Best for: musicians and AI-music channels, lofi and ambient visuals, lyric videos, and visualisers for podcast intros or soundtracks.

​
Step by step

1

Add the song

Song section with Custom, My Music and Library tabs and the audio drop zone

① Custom / My Music / Library · ② Drop or link your audio · ③ Browse all library tracks

  • Custom — drop an audio file (MP3, WAV, OGG, MPEG or M4A), or paste a YouTube, Suno or direct audio link and click Process audio.
  • My Music — a track you made with AI Music or uploaded before.
  • Library — 49 ready-made tracks (Browse all shows them with mood tags).
Once the song is in, trim it to the part you want — the video follows the trimmed length.
2

Choose the format and quality

Aspect ratio and AI image quality (Basic, Pro or Ultra), as in Common settings.
3

Write the visual story (optional)

Visual story section with idea chips, the description field and Generate visual story

① Idea chips · ② Describe the look · ③ Generate visual story

Describe how the video should look — mood, setting, characters, story arc — in up to 4,000 characters. Click an idea chip (Neon city at night, lonely drive, Golden-hour road trip through the desert, Dreamy underwater world, Retro 80s arcade nostalgia) or Generate visual story to have the AI write one that fits the song’s length.Leave it empty and the AI decides from the lyrics and mood. Add up to 4 characters to keep a protagonist consistent across scenes.
4

Pick the style and captions

A visual style and a caption style. Lyric captions are transcribed from the song’s vocals for free.
5

Choose the generation path and scene length

Generation path cards and the scene length setting

① Scene images · ② Complete video · ③ Let AI choose scene length

  • Scene images (cheaper) — one image per scene with smooth cinematic motion. Cheaper and faster; bring any scene to life with video later in the editor.
  • Complete video — every scene is animated into moving video and the music video is finished automatically. The most cinematic result; uses more credits.
  • Let AI choose scene length — short 6-second scenes, the most images for the best flow. Turn it off to set one exact length between 6 and 15 seconds (longer scenes = fewer images = lower cost).
6

Style the waveform (optional)

Waveform settings with Bars, Line, Wave, Dots, Radial and Area styles

① Audio waveform overlay · ② Style · ③ Position and size

Turn on the Audio waveform overlay to draw a reactive waveform of the song over the video:
  • Style — Bars, Line, Wave, Dots, Radial or Area.
  • Position — bottom or center; Size — S, M or L.
  • Opacity, Glow (a soft neon halo) and a start/end color gradient.
It reacts to the real audio, so beats and drops are visible.
Click Create Music Video. The estimate appears as soon as a song is added and depends on its length, the scene length and the path you chose.
Lip-sync for singing characters is coming soon.

​
Tips

  • Trim to the best 30–60 seconds for Shorts — the chorus usually works best.
  • Describe a journey in the visual story (from a rainy city to a sunrise on the coast) so scenes progress instead of repeating.
  • Use your own or AI-made music to avoid copyright claims when you publish.
Lofi channels love this format — try the Lofi music visuals Autopilot template.