Why would you ever want to remove vocals from a song? Well, let’s say you’re having a karaoke party and you absolutely need a version of Funky Cold Medina with a little more bass and a greatly toned-down Tone Loc. Or maybe you’d really love the music behind Sisqo’s Thong Song if it weren’t for the “singing” part of the song about thongs. Nowadays, you can apply a little technology to so many problems, and presto: they’re solved! If humankind can close in on the search for the elusive Higgs Boson particle, take vivid pictures of distant galaxies, conquer smallpox, prepare to clone woolly mammoths, and brave the delicious perils of mixing chocolate with peanut butter, then removing vocals from a song is comparative child’s play. (Fair warning: Just don’t try to remove peanut butter from my chocolate, or you might return from the experiment minus a hand! Happy holidays to you.)
A while back, Daniel from Live Tech Australia was kind enough to give us a step-by-step tutorial about how to remove vocals from a song using the Audacity freeware. We thought it might benefit the LockerGnome community to have the steps in writing to go along with the visual. How does that grab you?
First, of course, you need to download the version of Audacity that plays nicely with your system (it’s available for Windows, OS X, Linux, and Unix). If you don’t know what Audacity is, it’s “free, open source software for recording and editing sounds.” What? Sound editing software that doesn’t cost a zillion bucks? Yes, it’s quite an audacious concept, I agree.
Once you’ve done this, open the song file that you’d like to edit in Audacity. The software will then import it, and you’ll be able to see a visual representation of its sound waves. Fancy!
Open up a stereo track by clicking on it.
Click the little arrow to the left of the track, and scroll down to Split Stereo Track in the dropdown menu.
Select the bottom track.
At the top of the screen, click Effect, then scroll down to select Invert in the dropdown menu.
Audacity will chew on this command for a few seconds, bringing up a progress bar that won’t even be visible long enough for you to go take a restroom break. Fast!
Select the bottom track again, and the little arrow to the left. Scroll down the menu and select Mono.
Select the top track, the little arrow to its left, Scroll down the menu and, again, you’ll select Mono.
Now, with any luck, you should be able to play back the modified ditty sans vocals. If it just plain doesn’t work, there could be number of reasons why — and it has nothing to do with Audacity’s personal vendetta against you for exploiting its generosity in being free. (Don’t be silly! Software has no feelings.)
While there is professional software out there that can more likely accomplish this task more cleanly, it can cost a pretty penny (or even a nice nickel, a dishy dime, or a coquettish quarter). Then again, while this method is pretty quick and painless — as well as free — it’s not guaranteed to work for every song out there. Even if it does work, you may find the resulting quality to be a bit lackluster. Short of having the master, multi-track tape of the song in question and a mixing board at your disposal, what you’re trying to accomplish may simply not be possible. The human voice is a many-splendored (or not-so-splendored) thing, and some voices just blend in with the background music at a pitch that matches too closely for the purposes of this. As a result, you may have excellent luck with extracting vocals from one of Tiny Tim’s more… shrill musical numbers, whereas trying to do the same with an Enya lullaby may be pretty much impossible.
If all else fails, you can always look up the song you want to hear without vocals at a karaoke DJ warehouse or something. It’s startling how many versions of Journey songs you can find in any such catalog (and have no fear — they’re still at least as horrible as the originals; don’t stop believing, forsooth). These are generally re-recorded by studio musicians for the very purpose of karaoke (and you thought karaoke had no purpose!), so you won’t get the commonly garbled or muddier sounds that often show up in the above mentioned, vocals-removed-from-the-original method.
In any case, if you can’t remove vocals from a song using Audacity in the way that Daniel describes no matter how many times — or how many different songs — you’ve tried, perhaps you’ve since stumbled upon (or invented) a better way to get the job done? Please share it with the rest of the class in the comments below so that we may all may be enlightened by the fruits of your curiosity. Thanks!