When asked about digitally processed or synthesized vocals, the average person might reference Kanye West or Rihanna’s use of auto-tune, or perhaps talkbox vocals from the likes of Peter Frampton. Maybe the casual listener even knows about Kraftwerk’s “The Man Machine” or Air’s frequent use of vocoders, but they very likely have heard the vocoded lines from Tupac Shakur’s “California Love,” delivered by Roger Troutman (more on him below).
When used wisely, synthesized vocals can add an additional sonic character to a track, regardless of genre. Arca and The Knife’s Karin Dreijer Andersson, in particular, are two artists who have exploited the possibilities of vocal synthesis. In last year’s single “Reverie,” Arca warped his operatic vocals in an alien-like way (starting at 2:20), while Dreijer Andersson regularly uses pitch and formant shifting to make her voice sound low, dark and unsettling.
The techniques of vocal synthesis and processing are many and varied. And the real fun comes about while experimenting to find new sounds, not trying to exactly replicate a synthesized vocal timbre. But, for those looking to get started, what follows are three ideas for processing vocals with VocalSynth.
1. Create ethereal vocals
Want something more ethereal? Well, there are plenty of ways of thinking about how to do this. The ultimate approach is really up to the individual as far as what sort of timbre, effects, and overall sonic aesthetic is desired. But let’s say you want to create a sound somewhat inspired by Grimes or, heck, maybe even a more digital sounding version of Enya!—that is, breathy, dreamy high notes. Well, now is a perfect opportunity to flex the VocalSynth muscles with various modules and other features.
Grimes’s song “Colour of moonlight” at 1:29 has a really interesting layering of effects—chorus, reverb, and perhaps a little bit of a vocoder effect. The sample we have here—an “Ahhhhh” taken from Splice—is split into two tracks. One track is in G# in the Major scale, while the other is set to F# in Major.
Track 1 uses Polyvox, Compuvox, and Vocoder. In the Polyvox module, the vocals are Formant shifted up to 2, while Character and Humanize are set to 10 and 73, respectively. Shifting Formant brings the vocals up into more of the Grimes range, while the Humanize setting adds more digital color to the sound.
Adding even more digital flavor are Compuvox and Vocoder. In this loop, Compuvox is set to Basic Syntax and Math, while Vocoder is set to Soft Wonder and Smooth. If you want only a slight robotic vibe, keep Compuvox’s settings relatively low; though you can be a bit more aggressive with Vocoder’s settings if it’s set to Soft Wonder. The Filter and Delay effects also help add some more sonic color, so experiment with those in conjunction with the above (or even all four) modules.
Track 2, set to F# in Major, features a more minimalist approach. It only uses Polyvox, and this engine’s settings are identical to the ones in Track 1. While Track 1’s Voices were set to a 7th, a suboctave, and a superoctave—for experimental purposes—Track 2’s voices are set to a 5th, a suboctave and superactive, all of which are cranked.
When combined, these tracks unite into the heavily layered “Ahhhh” that would fit on something like a modern electronic tune. It could even be used to create something like Air’s dreamy, synthetic-sounding vocals on “Run,” each track set to a different octave, which itself was inspired by 10cc’s classic track “I’m Not In Love.” (Note: To get the “Run” or “I’m Not In Love” vocal sound requires tons of multitracking, so research how Air and 10cc did it.)
2. Darken your vocals
This loop features a sweet but melancholic vocal from Roniit Silk, downloaded from Splice. Using VocalSynth, Roniit’s vocal is stripped of some of its sweet, ethereal vibe, and augmented with a darker, more foreboding sonic hue.
First, the vocal’s root is set to G in the Minor scale, and the dry signal in the Voices section is dropped to zero. Since this, along with a Formant shift in Polyvox (even with each of the three voices in the Voices section in suboctaves), doesn’t darken Roniit’s vocals enough, a touch of Compuvox (set to Your Base and Math) is added to the mix. This combination gives the vocal a tonal quality somewhat reminiscent of Karen Dreijer Anderson’s vocals for The Knife (See: “From Off to On”) and her Fever Ray project.
In and of itself, this sound is definitely cool, but not really what we are going for. To add some sonic complexity and grit, activate Filter section, set it to New York LP, and then open up the Filter to 4807 before adding a bit of Resonance (setting it to 8 to 20 does the trick). You will want a wet signal to achieve something like (or beyond) Karin’s sound, so set Wet/Dry signal to 65 Wet or up. Even more grit is added with the Distortion effect, which is set to Analog and Drive.
Want to play around with the sound some more? Activate Transform, set to Radio, then peg the Width to the middle, and tip the Wet/Dry signal just a smidge to the Wet side. With this as your baseline sound, you can then begin experimenting with the other three modules, and you will eventually arrive at some pretty wild sounds that go far beyond The Knife and Fever Ray.
3. Turn your vocal into a complex synth
This acapella vocal sample, supplied to iZotope by Montreal-based performer and multimedia artist Annie Sama, is a great example of how VocalSynth can be used to turn a voice into a multi-hued synthesizer. In this sample, Annie, who uses vocal effects in her recordings (check out her 2016 single “Polyday”), delivers a pretty rhythmic vocal, completely free of processing. Her voice almost has a percussive quality, with its combination of singing and rap delivery.
To process Annie’s vocals with VocalSynth, each of the four synth engines were activated. In Polyvox, her vocals are Formant shifted down to -2, with Humanize set to 68 to give her voice a sort of otherworldly quality. The Vocoder’s Detuned Dream, set to Vintage, adds a sort of robotic shimmer to Annie’s voice, which is further augmented by a touch of Compuvox’s Elliot’s Modern setting, and Talkbox’s Noisy Analog, Formant shifted down to -6. The Filter and Delay are also activated, which further sculpt and add depth to Annie’s vocals.
Most of the vocal color in this loop is supplied by Polyvox and Talkbox, mixed with the Dry signal. Other tonalities are added through through various octaves in the Voices section, which can be seen in the screenshot.
Annie Sama is set to release her new single late February 2018.