I’ve been there—wanting to make a drum cover video on the latest song only to find that no drumless version exists. Is there a way to isolate and remove a drum track with ease? Today I’ll explain a few ways you can take the drums out of any song with just a few clicks.
Table of Contents
What Are Drumless Tracks?
I’ve always searched for drumless songs of popular music online in my twenty-plus years playing the drums. Drumless versions are great for making covers or learning your favorite band’s music. YouTube is a fantastic resource in this regard, along with the various websites that spawned out of drummers seeking play-along tracks from their favorite artists.
These websites generally relied on obtaining the master recordings, whether by contacting the producer/label or finding the stems stored away in the raw files of the latest Rock Band or Guitar Hero video game. There were ways to remove drums using VST plugins within a DAW, but each instrument’s isolation was shoddy at best.
How Do You Remove Drums from Songs Today?
Today, however, technology is far superior. We now have complex AI algorithms that can detect the tonality and frequencies of the audio spectrum. And they can isolate to an astonishing degree. There are two ways you can go about removing drums from any song:
- Use a web-based application (no software needed)
- Use your digital audio workstation (DAW) with a VST plugin
Which you choose is ultimately up to you, but I like to use web-based apps, as I think they’re easier to use and, in my experience, faster.
Remove Drums From a Song Using a Web App
If you’re like me, you don’t want to waste time trying to find audio files, convert them from iTunes, bring them into a DAW, and then finally remove drums from a song. There’s too much clicking, file management, and extra work.
I prefer using a cloud-based AI application called Moises.ai. Moises allows you to extract the drums from any public media URL, whether SoundCloud, YouTube, etc.
The UI is simple and easy to use. Inside the dashboard, you can keep a library of all the songs you’ve isolated for easy playback and downloading.
Some tracks don’t isolate the greatest, but I haven’t had any issues with removing drum parts for the most part. Once separated, you can download individual .mp3 files of the vocals, drums, bass, etc. As of now, you can only download .mp3 files of separated tracks unless you pick .wav, which is a bounce of the levels from the mixer.
From here, I bring the songs into my DAW to play-along. The web app makes it super simple to create drum covers fast. You can also play-along within the web mixer, if you like.
In the example above, I have a track, Kid Charlemange by Steely Dan, with the drums muted.
Moises has a free version and a paid subscription. The free version allows you to try the service out, but you won’t be able to save any of your songs in their library, and songs are limited to 5 minutes in length. You can only upload five tracks per month, as well. The paid version is well worth your money.
Other solutions include:
Removing Drums from a Song Using VST Plugins
If I didn’t convince you above and insist on using VST plugins to remove drums from your favorite music, here are a couple of options.
- XTrax Stem 2
- UNMIX Drums
- Drum Extract
The inability to customize or adjust settings when extracting drums from a track is one of the main downfalls of using a web-based application. VSTs offer a bit more flexibility and potentially could deliver a better result.
If you can’t get a good result with either solution, you can also try to EQ some frequencies out of a track to lower the drums’ perceived volume. Using a parametric equalizer in your DAW, try reducing these frequencies a few dB (these aren’t exact numbers, YMMV):
- 150 Hz (snare drum)
- 90 Hz (kick drum)
- 4k (cymbals)
Play around with the EQ and mix your drum tracks above the recording to eliminate your kit’s flamming against the original recording.
That’s all for today! Have you tried any of these techniques? Let me know if I missed one, and I’ll be sure to add it to the list. Thanks for reading.