This is a copy. Use Snipmill at snipmill.com.

Skip to content

Take the singing out of a song and keep the music, ready to sing along.

Settings

Settings

KeepThe music, for karaoke
Runs on Snipmill's server: only the song is sent, and deleted after. How we keep it private

Good to know

What the karaoke track holds

Drums, bass and every other sound the model doesn't take for voice, mixed back together at the song's full length, as a 48 kHz stereo MP3 of about 1.4 MB a minute. A 3-minute song makes a karaoke track of about 4.3 MB. For the words to sing along to, choose "Vocals only" after it: the song isn't sent again.

From the track to a sing-along video

A karaoke night or a practice clip needs the words on screen. You can build that from the tools here, one after the other:

  • Save "Vocals only" as well as the instrumental, and open the vocals in Audio to Text. Clean singing is easier to write out than the full mix, but check every line: a speech model can mishear sung words.
  • Fix the lines against the song and save them as SRT. The times fit the instrumental, because nothing was cut or moved.
  • Open the instrumental in Audio to Video to get a video with a picture behind the music.
  • Open that video in Add Subtitles with your SRT, so the lines are drawn into the picture.

What goes to our server

Only the song goes to Snipmill's server, when you press the button that starts the job; the line under it says so. The server deletes it once it has answered and keeps no copy. The instrumental comes back to this page, which saves it as an MP3 in your browser.