Creating Multitracked Stacked Vocals with Mixcraft
By Mitchell Sigman
Ever wonder how to get huge, lush-sounding backing vocals? I used to think the secret was simply being a stellar vocal talent, but fortunately, that’s not the case. The real secret is mainly just stacking a lot of tracks for a smooth, “homogeneous” sound. And as you’ll see, you don’t need to be a great singer.
I find that stacking up four layers of each harmony note works well. That’s enough to make things sound real big; I find any more is sort of diminishing returns. This means that if I’m tracking a three-part harmonized backing vocal, I’ll end up with a total of twelve tracks, i.e. four unison passes of each line. My method for recording myself used to be as follows: I’d create a “recording” track, then a bunch of empty “playback” tracks beneath. I’d hit the record button, run to the mic, sing one pass, lean over, hit the stop button and if the take was good, I’d drag it to one of the empty “playback” tracks. This works ok, but as you can imagine, you do a lot of starting, stopping, dragging, deleting, running-back-and-forth, etc. If you’re familiar with Mixcraft’s “take mode” and loop recording, there’s a much better way… here’s how:
When a section of a song is set to loop, Mixcraft automatically records a new pass with every “go-round” of the looped section – each pass is a separate audio recording and is assigned to a new automatically created “lane” on the record track. The most common use for take recording is to lay down multiple passes of a vocal or instrumental passage for later editing into a single optimized performance, but there’s no reason we can’t use take recording to build multiple vocal tracks, with the intention of using all of them (or at least most of them, depending on the quality of the performance).
Take recording mode is automatically enabled when a loop section is defined; here’s a quick clip showing how to loop a section (click the “full-screen” button at the bottom right corner of the movie to see more clearly):
Now that we’ve looped a section, here’s a video showing the recording of my three-part harmony (with four layers of each line). The idea was to end up with an acapella vocal, but I tracked an electric piano part for pitch reference. I edited some of the tracking out, because watching me sing the same lines over and over is little lengthy, but I actually did record four layers of all three harmony parts in one giant “take” (in other words, Mixcraft cycled around 12 times).
There’s a couple things I want to point out about the video. First of all, you certainly don’t need to record all the vocals in one take – you could separately record the four layers of each harmony. But keep in mind that besides saving time, take recording lets you lay down unlimited takes and sort through the good and bad ones later. And while we’re on the topic of good vs. bad… you probably noticed that my raw vocal performances don’t exactly give Christina Aguilera a run for her money, and that’s part of the point. Stacking vocal performances can yield really impressive results even if you’re not a great singer (though admittedly I was coming off four solid nights of gigging here ). I find that being as consistent as possible with pitch, tone, and phrasing (i.e. making words the exact same length) is more important than beautiful tone quality. And unlike being born with a lovely set of pipes, tight vocal doubling is a relatively easily learned skill. Another thing you might notice is that you can’t hear all the parts stacking up as I’m going. This is intentional on my part because hearing all the other parts distracts me a little from what I’m currently singing, so I have monitoring of previous takes off (hit the mute button on the record track). I’m sure some people would rather hear everything; if you prefer this, turn up the fader on the record channel.
In the video below, I explain how to move the takes to individual tracks, pan them for stereo, compress and EQ them as a group using a SubMix Track, and add a final reverb (as a disclaimer, at 6:18, I should’ve set the direct/effect mix knob in the reverb fully clockwise to “all effect”. Send effects should always be set fully wet, because you’re already getting the dry signal in the main mix). If your pitch is a little wonky, I recommend using multiple instances of an autotune-type plugin, set to the key of the song, for each individual vocal track (a single instance of a tuning plugin will not work properly on a SubMix track). Mixcraft’s GSnap Pitch Correction plugin is perfect for this. You’ll notice I used the TB Parametric EQ, which is included only with Mixcraft Pro Studio’s super deluxe array of plugins, but any EQ plugin could be used –even the basic channel strip EQ.
Below is the final, all singin’ (literally), all dancin’ (not-so-literally) audio result:
As always, questions and comments are welcome… happy singing!