This little vocal reverb trick is one that I don’t use all the time, but is fantastic for vocal performances that start off very quiet and intimate, but build into some kind of dynamic climax.
The technique I’m about to share with you is the one I used for the extremely dynamic vocal performance by Anikó Tóth on the song ‘Cast’ that I wrote in collaboration with Mike Sizemore. The song had appeared in print form in John Carpenter’s Tales for a Halloween Night.
The most common practice when adding reverberation to a vocal is to have the ‘verb on an auxiliary channel and use a send on the vocal track to the auxiliary channel to add reverb. The origin of this practice is largely based in the fact that an old analogue studio had a limited number of reverb units (perhaps only one, in a smaller studio), as well as a limited number of auxiliary channels. Using this technique, the mixer could use the same reverberation unit, on the same setting, for every vocal or instrument that needed it, if absolutely forced to.
The engineer could work around these limitations to some degree by printing a particular reverb to tape, but that would still mean the engineer was limited by the number of tape tracks. In those days, the track count was usually limited to 24, although towards the end of the analogue era, the track count may have reached 48, in some of the bigger studios.
Well, ok, but what impact did these limitations have on dynamic performances?
Think about how reverberation works in the natural world, such as in a cave or a large room with bare walls. If you speak or sing quietly in that cave or room, then only a small pressure wave will be created in the air particles, and, therefore, there will be limited reflections before that pressure wave runs out of energy. If you shout, or sing loudly, into that space, then you will create a much larger pressure wave in the air molecules that has much more energy and can bounce around off the reflective surfaces for much longer.
So, couldn’t we, then, just automate the reverb send on a vocal track to emulate this behaviour? If we send less of the vocal track to the reverb on the quiet sections and more on the loud sections, wouldn’t that be pretty much the same thing?
Well, sort of, and, in terms of music production, that is a perfectly valid approach that mixers use all the time. However, if you really think about it, that’s not really what the reverberation is doing.
When we sing quietly in the cave, the pressure waves have only enough energy to bounce off the surfaces very close to us before they run out of energy, but the waves produced by the loud singing has enough energy to reach surfaces much further away and bounce back, hitting all the other surfaces much closer to the us, before finally reaching our ears. Our brains don’t interpret this as more reverb, or less reverb, but as two distinct varieties of reverberation.
One very famous solution to this problem can be found in the recording of David Bowie’s vocals for the song ‘Heroes’ from the Tony Visconti/Brian Eno produced album Low. In short, Visconti was left with one track for Bowie’s very dynamic lead vocal. But, how could he capture the appropriate reverberations for the varying dynamics of the song? Visconti came up with a solution that was sheer genius in its simplicity and elegance.
He set up one mic close to the singer, one about 15 or 20 metres away, and a third one further away, still, in a very large room. Noise gates were set up on the channels for the two distant mics, which only opened when Bowie sang at a certain level, which is when they would capture the reverberation of the room. All three mics were, then, mixed down on the desk, and a dynamically nuanced performance, with appropriate reverberation, was captured on that final remaining track.
This can be a fun method to experiment with at the mixing stage by setting up three different reverbs on three separate auxiliary channels, each with a noise gate in front of the reverb plug in. In fact, you can buy a single plugin called TVERB from Eventide that emulates the whole set up used on the Bowie session as closely as possible.
You can buy a copy of this plugin at Plugin Boutique here.
(As a brief aside, there is a hilarious cartoon about Bowie, Eno and Visconti here)
However, this strategy starts to fall apart if you heavily compress the vocal before sending it to the reverbs on the auxiliary channels. If, as is common in modern pop and rock production, you smash the vocals with compression to get the performance nice and even throughout and to pop against the background, then you are going to run into problems triggering the noise gates. Either the level will be so high throughout that all the gates open, so you hear all three reverbs all the time, or none of them open and you get no reverb at all.
Is there a way around this?
Well, yes.
The method I prefer - and there are a number of ways you could do this - requires a completely uncompressed (or, at most, very lightly compressed) vocal track. Then, take this uncompressed track and copy it to two other tracks in your lead vocal group. The first one would be compressed as normal; the second would be your parallel compression track that can be smashed to billy-o, (assuming you use them (if not, why not?)). Both of these tracks would be routed to your lead vocal bus, and neither of them are sent to the reverb auxiliary, or auxiliaries. The third uncompressed track is treated differently. This track is not sent to the lead vocal bus (I usually have no output, or send it to outputs that aren’t in use), but has a send going to the reverb auxiliaries.
Now, your reverbs can react naturally to even the smallest fluctuations of level in the performance, creating, combined with send and level automation, a huge range of expressive possibilities.
One extra little trick I add is to put a compressor after the reverb to level it out a little. Remember, we are trying to replicate the sound of natural reverb with a dynamic performance, not swamp the entire vocal in ‘verb as soon as things get loud!
Another little creative trick can be to have the compressor, or even a second compressor, side-chained to the main vocal track, so that, not only are the lyrics not obscured by excessive reverb, but we get a wonderful “blooming” of reverberation in the spaces between the lyrics. Go nuts! Have fun with it!
Now, if you’re like me, you might like to compress a vocal through an analogue compressor on the way in; you might want to do a little routing jiggery pokery at the recording stage to get a compressed and uncompressed take. I’ll leave you to work that out.
The actual reverb I used on this track was a plate reverb, similar to this one. These are fantastic plugins that can be bought from Plugin Boutique, along with other great plugins, below.
