I often record double-tracked vocals and pan them left and right, but was listening closely to some other artists and noticed how the vocals seem to be close-up and spread across the stereo field, without appearing doubled up — for example, David Sylvian’s voice on ‘Wonderful World’ (from the Nine Horses collaborations). Do you have any recommendations for ways to achieve this? What I end up with sounds like it’s just lying centred on top of the lead vocal.
SOS Forum post
SOS contributor Mike Senior replies: Listening to the clip you’ve linked to, this doesn’t seem like any sort of double-tracking — it’s a single voice with some processing. The problem is that the processing is subtle enough that it’s difficult to be sure what’s going on. However, given that I can’t really hear any obvious modulation element in the effect, I’d hazard an educated guess that we’re hearing one or both of two very commonly used mix treatments.
The first of these is the classic stereo widening effect closely associated with a couple of hardware studio stalwarts: the old AMS DMX 1580S digital delay units and the early Eventide Harmonizers. The effect in question is a mono-to-stereo pitch-shifter–delay send effect, where both the left and the right channels are delayed by between 5ms and 20ms, but with one channel a few milliseconds ahead of the other. The earlier of the two delay taps is then pitch-shifted downwards by 2-10 cents (cents being hundredths of a semitone), while the later tap is pitch-shifted an equal amount upwards. The net result of this process is that the delay taps separate themselves from each other as far as the listener is concerned, and sit at the extreme edges of the stereo field.
Adding this widener effect to a dry vocal gives the impression that it is spreading across the stereo image, the width depending on the level of the pitch-shifted delays. Because the delay times used are so small, you don’t really perceive any kind of ‘echo’, but if the level of the delays increases there is the danger that they begin to phase-cancel noticeably with the dry signal and thereby change the overall vocal tone. If you find that this starts happening, try adjusting the exact delay times a little, as this will shift the phase-cancellation frequencies and allow you to find a setting that better suits the sound you’re after. That said, ever since the band Chicago went out of fashion, few people without permed mullets use more than a touch of this effect, so tonal changes are unlikely to cause much of a problem for you if you’re after that kind of low-key David Sylvian sound.
The second effect that I think is at play here is a very short stereo reverb/ambience or early-reflections patch — so short, in fact, that there’s no hint of any reverb tail at all. (There is a more audible reverb on that song as well, but I suspect that’s a separate effect again.) Using this kind of reverb at a fairly low level and with very little pre–delay causes it to meld with the vocal sound itself, rather than being heard as an effect in its own right, and you can then use its stereo width to your advantage, perhaps adjusting it to suit your needs using your return-channel stereo balance control or a plug-in such as Voxengo’s MSED. As with the short delays in the Harmonizer-style effect, there is the potential for this short reverb to phase-cancel with the dry sound somewhat, so careful adjustment of the distribution of the earliest reverb reflections can pay dividends. However, it’s also worth mentioning that a change in overall tone can be used to advantage in many cases. For example, unusually bright reverbs are often applied to add a kind of extra high-frequency enhancement.
With a combination of those two effects (Harmonizer-style stereo widener and super–short reverb), I reckon you could probably get very close to the sound you’re hearing in that track.