Here’s a nifty little trick that I use live all the time. I’m sure I’m not the inventor of this method, but what the heck, I’ll name it after myself anyway!
This is basically parallel (New York) compression with a twist, so here’s a basic rundown on where it all started. (If you’re already hip to the parallel compression band-wagon, skip ahead to the next paragraph). Parallel compression is basically making a duplicate of your original, unprocessed signal, and applying compression to that copy only. This has many uses, and many benefits compared to just adding compression to one channel alone. We all know that adding compression turns down the peaks of a given signal, thereby limiting the dynamic range of that signal. However, the negative is that every time the compressor hits, it’s turning down the whole signal, not just shaving off the peaks. When overdone, this can sound very awful and will leave a bad taste in most mouths. You loose all the punch and attack of the instrument and things tend to turn out lifeless after a massive dose of compression. Enter this parallel duplicate!!! You get the original, unaltered signal with it’s full dynamic-range-glory, and when you slowly blend in the heavily compressed track, you start to hear the energy and low-level stuff the compressor brings out. Best of both worlds! (Many people claim the difference between this and New York compression is that in NYC they add some top and bottom EQ to the compressed track too. Take it or leave it)
So, here’s where Joe’s Auto Compressor (JAC) comes in!
I use this mostly on the lead vocal, when the singer regularly goes from a quiet whisper, to a full-on lung release all night long.
Start with your main vocal channel, EQ to taste, send to your FX etc. Maybe even add a touch of compression here, but no more than 2:1, knocking only 2-3dB off the peaks. This channel is now your “Yelling” fader.
Next, duplicate this to the next channel over (either digitally or with a Y cable, if digitally, make sure the same processing is applied even if left flat to ensure the signals are time-aligned).
To this channel, add stupid compression. I’m talking 20:1 or more, fast attack and release, get wild. Here’s where you play it by ear, but I tend to let that compressor take off between 10-20dB when the person is singing loudest. For example, lets go with 15dB. Lastly, add 15 of makeup gain to this channel, and name it “Whispering”
Set the faders to the same level, and let the compressors do the mixing!
So here’s what happens:
Singer is singing quietly, the same signal is hitting both channels, so the one with 15dB more gain on the output is the dominant one you hear. As the singer starts belting it out, the compressor on the “Whisper” track ducks that signal out of the way, and you hear more of the “Yelling” channel. They won’t swap completely, but the levels will be closer, and you end up getting a good blend of the uncompressed signal automatically, which sounds infinitely more realistic than hearing full compression all night long!
While these techniques can be used live or in the studio, I find this one most useful in a live setting. (Just be sure that neither of these signals are the ones feeding monitors. That much wonky compression can really throw off a singer. Just make a 3rd duplicate and use that for wedges only if you’re driving them from FOH)
Give it a try, mess with those output gains, mess with the threshold, mess with your lighting guy 😉