Reverb can be a double-edged sword in production. While it can add ambience and a third dimension to a track, it can also make it sound less clear and discrete. Fortunately, there’s a way to keep the good while ditching the bad and the ugly.
In this article, I’ll show you how to use ducked reverb to add depth to vocal tracks without gumming up the works. I’ll also demonstrate radical uses of the technique to create outrageous effects. I’ll point out a plug-and-play product that produces ducked reverb, but I’ll also show you how to design your own ducked reverb patch from plug-ins you likely already own. But first, what exactly is ducked reverb?
NOW YOU HEAR IT, NOW YOU DON’T
Ducked reverb automatically lowers its output when its input signal exceeds the threshold level you set. Slap it on a vocal track and set a low enough threshold, and ducked reverb will decrease in intensity whenever the vocalist sings, drying up and clarifying the track. At the end of vocal phrases, when the singer’s level falls below threshold, the reverb’s output will regain full volume to create an enhanced sense of space.
Fig. 1. The Waves H-Reverb plug-in uses controls situated in the lower-left corner of its GU to produce ducked reverb. The Waves H-Reverb plug-in does the heavy lifting for you (see Fig. 1). Simply select Duck mode in the GUI’s Dynamics section, and lower the Threshold control until you hear the reverb’s output decrease during vocal phrases. The Recovery control governs how quickly the reverb will recover to full output during any gaps in singing. Make sure the plug-in’s reverb time is long enough—and Duck mode’s recovery time fast enough—that the tail is still decaying after the recovery time has elapsed; otherwise, the effect will already be finito by the time the plug-in’s output level recovers. Because HReverb lacks an input control, place it on an aux and bus the vocal track to it (vs. placing the plug-in on the track’s insert); you’ll often need to use the bus’ send to lower the vocal track’s level far enough below the plug-in’s threshold to achieve deep ducking.
ROLL YOUR OWN
You can produce ducked reverb using plug-ins you probably already own. Bus the vocal track to a stereo aux. On the aux’s inserts, instantiate a compressor plug-in followed by a reverb plug-in. The best compressors to use for this application offer threshold and release-time controls and provide high ratios, very fast attack times, a hard knee and lookahead detection.
Fig. 2. FabFilter Pro-C 2’s comprehensive control set makes the plug-in an eminently capable ducker for reverb placed downstream. The FabFilter Pro-C 2 compressor plug-in fits the bill perfectly (see Fig. 2). Dial in the fastest possible attack time, the hardest knee, an infinity:1 ratio, and 60dB range. Using about 10 ms of lookahead detection prevents transients at the start of vocal phrases from spiking the reverb before the compressor’s attack time has completely elapsed and peak gain reduction is achieved. Lower Pro-C 2’s threshold until reverb is attenuated the desired amount when the vocalist is singing on the track. Set the release time to be fast enough that the compressor’s output level fully recovers (producing no gain reduction) in the gaps between vocal phrases and before the downstream reverb’s tail has completely faded; after the compressor’s release time has expired, the reverb’s input will fully receive the vocal track’s send signal at its set level, causing the reverb’s output level to blossom.
If you set the compressor’s release time too short, the reverb may bloom during very brief gaps between sung lyrics within a vocal phrase. Conversely, if you set the compressor’s release time too long, the reverb may not bloom enough before the following vocal phrase causes it to be attenuated once again. Keep in mind that the lower the threshold and the higher the gain-reduction range you set, the quicker the release time will need to be in order to restore the reverb’s level enough that the reverb’s bloom will be readily discernible. The reverb’s tail must also be longer than the time it takes the compressor’s release to restore signal level. When you find the sweet spot for all these controls, the reverb’s output level will decrease during singing and rise at the tail end of each vocal phrase.
The best conventional compressors will duck reverb only so much, attenuating it perhaps 30 dB but not eliminating it. To completely dry up a vocal track during singing, you need a compressor that offers negative ratios. With negative ratios selected, the higher the compressor’s input signal, the lower its output level becomes.
Fig. 3. The elysia mpressor’s negative ratios can be used to create extreme duckedreverb effects. The elysia mpressor, a hard-knee compressor, is the only compressor plug-in I know of that offers negative ratios (see Fig. 3). The mpressor can produce extreme ducked-reverb effects in which loud, bone-dry vocal phrases end with exploding reverb. (Make sure you turn off mpressor’s GR Limit switch—deactivating a range function—to allow the deepest possible gain reduction.) The mpressor’s only shortfall in this application is its omission of lookahead detection, which results in transients at the start of vocal phrases sometimes triggering reverb bursts. Fortunately, the mpressor includes an external-sidechain input, which you can use to make the plug-in respond as if it is using lookahead detection. To learn how, please read my article “DIY Lookahead Detection” in the June 2015 issue of EM (available at emusician.com).
Whether used to clarify vocal tracks or create explosive effects, ducked reverb is a powerful weapon every creative engineer should have in his or her arms depot.