free, no sign-up, nothing leaves your browser

Make a music video

Drop in your song and pick a look: equalizer cities battling across a night lake, glassy VU bars in a mirror pool, a vaporwave drive between mountains made of your song, or The Starry Night pulsing in 3D. Paste your Suno lyrics and it knows who's rapping; add timed captions and the words land on the beat. Then download it as an MP4.

1. Your song Choose an MP3 (or WAV, M4A). Up to 8 minutes. It stays on your computer.

2. Who raps when (optional) Lyrics, as you gave them to Suno (tags like [Verse 1: Name], [Male Vocals], [Female Vocals] and [Chorus] decide the side)

Load lyrics (.txt) Load timed captions (.lrc, .srt, .vtt)

No lyrics? It still works: it finds the sung parts and gives the lower voice to the first side and the higher one to the second. You can fix any line on the timeline.

3. The look

EQ BattleEqualizer cities across a night lake, fireworks on the punchlinesVU LakeGlassy VU bars in a mirror pool, clouds, mountains and fireworksVaporwave HighwayA neon drive to a striped sun between mountains made of your songStarry NightVan Gogh in 3D: the stars, the moon and the windows pulse to your songPixel StageA 16-bit concert: LED wall, sweeping beams, pyro and a bouncing crowdScope TerminalA green CRT: your song on an oscilloscope, the lyrics typed into a log

Two voices (a battle)One voice

Title Subtitle

First voice Colour

Second voice Colour

Show the lyrics on screen

Your video shows up here.

0:00

Quality

This is a demo: Fast Food Fight by the Sloptrawler. Press play and try the looks, then clear it and choose your own song.

How it works

The looks are code, drawn on your graphics card in your browser. EQ Battle is the engine from the Sloptrawler's rap-battle videos: each city is an equalizer whose towers are frequency bands, lit while its MC raps. VU Lake rebuilds the Sloptrawler's VaporVU visualizer. In Vaporwave Highway the mountains are a rolling 3D spectrogram of the song, one range per voice. Starry Night lifts Van Gogh's painting into depth and makes every star, the moon and the windows pulse with the band under them. Pixel Stage and Scope Terminal put the same song in a 16-bit concert and on an oscilloscope. Every look splits the song into its centre (mostly the voice) and everything else, so whoever's rapping leads, and fireworks or flares land on new sections, chants and lines that end with ! or ?.

Who raps when comes from your lyric sheet's tags. Timed captions (.lrc, .srt or .vtt) put each line on its time; without them the lines are spread over the sung parts, which is close but not exact. Lines nobody tagged go to the lower or the higher voice.

The video is drawn frame by frame on your graphics card, so a three-minute song takes a few minutes at 1080p. Keep the tab open while it runs. It works best in Chrome or Edge on a computer.