In “Sound Design Workshop: Off the Wall” (see the August 2005 issue of EM), I discussed some of the creative possibilities of slicing, displacing, and offsetting segments of the wet-only portion of a reverb-processed drum loop. This month, I'll start with a reverb-processed vocal track and process several copies of the wet-only version on parallel tracks. I've used Logic Pro 7 for these examples, but you can do the same things with any multi-track DAW and a basic set of plug-ins.
FIG. 1: With the use of high- and lowpass filters, the frequencies of the wet file can be split across several tracks. Here, they are also panned and offset. The center track is being processed with a tape-delay plug-in.
I've processed the vocal track with a straight-forward plate-reverb setting using a 2.6-second tail (see Web Clip 1). One of the most basic examples of parallel processing is applying different EQ, compression, and other effects to the dry and wet files. That alone offers considerable flexibility and control in shaping the sound.
You can achieve effects that are even more unusual by splitting the wet file into several frequency bands for independent processing. For example, place copies of the wet file on three separate audio tracks, and place a lowpass and a highpass filter in series on each track. That gives you variable-bandwidth bandpass filters with which to isolate the desired frequency bands. You can then pan and slightly offset the frequency bands to thicken the sound (see Fig. 1 and Web Clip 2).
Once you have split and panned the wet-file frequency across several tracks, you can try a variety of additional things. You can add a filter with some LFO automation to one or more of the wet tracks; you can slice up the wet file and stagger it across several more tracks to create a panned tremolo effect; or you can apply heavy compression, overdrive, or distortion to just one of the tracks.
You can achieve further sonic variation by processing the wet file destructively. For example, you can reverse a copy of the wet file and play it on a different track, with the two copies panned slightly apart. Another trick is to use different pitch- and formant-shifting on several copies of the wet file (see Web Clip 3). Those techniques aren't for everyday use but, applied sparingly, add interesting sonic color.
Cut and Dried
Another creative option is to process the wet file with a gating plug-in triggered by a sidechain input. You can use any sort of material from either the same song or another source for the sidechain input. For example, you can use one of the song's rhythm tracks, so that the gating of the wet file is not rhythmically tied to the phrasing of the original vocal. Percussive keyboard parts and rhythm-guitar tracks make good sources, as well.
Instead of a gate, you can use a compressor-expander with a sidechain input. Using expansion (with the dry track as the sidechain input) produces an inverted ducking, in which the volume of the wet track rises and falls with that of the dry track.
A third alternative is to use an enveloped filter such as Logic's Autofilter plug-in to process the wet track. An envelope follower, which can follow either the signal being processed or a sidechain input, is typically used to trigger the envelope that controls the filter's cutoff frequency. The range of possible effects depends on the filter parameters available. In the case of Autofilter, an LFO whose amount is controlled by the envelope follower can also be used to modulate the filter.
Of course, you can combine those techniques, as I have done in Web Clip 4. For example, you can apply the gating to a pitch- or formant-shifted version of the reversed wet file; you can gate the dry file; or you can remove the dry file altogether. In the end, your ear (and the singer) must be the judge, but there's a wide range of vocal variations out there for you to explore.
Eli Krantzberg is a Montreal-based drummer, bandleader, and home-studio owner. Special thanks to singer-songwriter Angela Latham for the vocal examples.