Pitch Perfect: A primer on live sound pitch correction - Part 2

Originally published in Worship Musician magazine, September 2024.

There is a funny comment often attributed to Woody Allen that goes something like “I took a speed-reading course and was able to get through War & Peace in about 20 minutes. It involves Russia.” I feel a bit like Woody Allen in this part of our overview on pitch correction in relation to formants. The topic is quite deep and I don’t have nearly the space here to really get into all of the technical details. Formants don’t involve Russia, but they are really important to at least generally understand if we are to put high-powered pitch correction tools to work in live sound applications. So with that caveat in place, let’s dig into the very high level basics that we’ll need as we move ahead!

Alvin and the Chipmunks

One of the first albums I had as a child was Chipmunk Punk. If you’ve never heard Alvin and the Chipmunks sing My Sharona, you are missing out. Perhaps the best way to understand formants is to reference the Chipmunks. Recording a male vocalist and doing a strict pitch correction up an octave will make the vocal track sound like a cartoon chipmunk. If we wanted to create the sound of that male vocalist actually sounding like he was that human singing up an octave, we’d need to account for the unique formants of his particular voice. Although I’m mixing metaphors a bit, one way to think of a formant is like the sonic fingerprint of a given source. The fancy technical definition (via Wikipedia) for formant is “the broad spectral maximum that results from an acoustic resonance of the human vocal tract.” As we all know from our work on adjusting the EQ of different sources, there are certain frequencies that tend to be readily excitable and accentuated or attenuated across the sonic spectrum. If we simply double the frequency of a source without accounting for the particulars of these ‘acoustic resonances’, we end up with chipmunks from human voices. If we include some formant ‘corrections’, we can add back some of the variables that are in play for a given source and end up with something that is more human-like and less obviously tweaked.

the broad spectral maximum that results from an acoustic resonance of the human vocal tract.

As Woody Allen highlighted, there is probably a lot more to this that is necessary to study in order to really draw out the technical details. I recommend Perry R. Cook’s PhD dissertation with the snazzy title ‘Identification of control parameters in an articulatory vocal tract model, with applications to the synthesis of singing’ as a good starting point if you are really curious (1990). But even without this technical deep-dive, we can turn to a few different example scenarios to give us a feel of how we might use formants to our mixing advantage.

Let’s imagine that you have two male background vocalists in your worship band and need to use pitch correction to thicken up the sound and artificially create a ‘female’ vocal line in the mix. As we’ve noted, you probably don’t want a chipmunk to join the praise band. Sure, they’ll work for peanuts—but their particular vocal qualities aren’t likely what the worship leader had in mind. In this case, we can adjust the formant on the generated octave up ‘correction’ of the male background vocalist to sound more like a female singer. We could also run this in the other direction if we had female vocalists and wanted to weave in something closer to the sound of a male tenor voice. To get a better sense of this effect in action, visit the Allen & Heath? Vocal Processing webpage and play with some of the wet/dry settings on the audio examples: https://www.allen-heath.com/hardware/dlive-series/rackultra-fx/vocal-processing/

Putting it all Together

As noted in Part 1 of this series, there are a number of plugin suites available that offer this sort of formant correction. Allen & Heath offers a number of native pitch correction effects within the latest dLive V2.0 update (via the RackUltra FX card) to insert vocal processing in the console in realtime. As the website explains, “Vocal Shifter radically alters vocals with a formant control and the ability to shift the vocal up or down by up to 12 semitones. Vocal Shifter can produce anything from pitch doubling to the unreal robotic and ageing effects heard in EDM and other electronic genres.” This ‘shifting’ effect includes a formant control so that you can decide how you’d like to color (or ‘age’ or ‘robotify’) the signal according to your needs. “Vocal Tuner is designed for natural sounding pitch correction, with minimal audible artefacts, suitable for all genres of music, and any situation where the performer may need a gentle helping hand.” This more simple tuning effect doesn’t include the complex formant parameter adjustments and as noted is more of a ‘minimal’ helper for really subtle pitch correction in more of an exposed solo (perhaps lead vocalist) scenario. “Vocal Gridder produces faster, less natural-sounding pitch correction – as heard on countless hip-hop, pop, trap, and modern country productions – where notes “snap” to the target pitch almost instantly.” We’ll get more into the other pitch correction concepts in Part 3 of this series, but if Vocal Shifter is more about the ‘shape’ of the signal frequency formants, Vocal Gridder is more about the temporal (timing) aspects of the corrections. None of these descriptions are hard and fast rules, but hopefully they can serve as a good starting point to get our heads around pitch correction overall.

In short, if you don’t want a scurry of chipmunks up on the stage on Sunday, you’ll need to understand formants. And yes, a group of chipmunks is called a scurry. Now I need to dust off the ol’ Chipmunk Punk album and get out my dancing shoes—see you next month for Part 3 as we dig into the fun topic of pitch correction in live sound.

Originally published in Worship Musician magazine, September 2024.

By Jeff Hawley

Jeff Hawley currently heads up the marketing for Allen & Heath USA. He is pursuing a PhD in philosophy at York St. John University and serves as the Communications Officer for the British Postgraduate Philosophy Association (BPPA) . Jeff has a diverse background as a musician, engineer, and trendsetting industry executive. In addition to performing and producing everything from noise music to disco and Afro-Cuban jazz to avant-garde klezmer, Jeff has designed award-winning music industry products and accessories and directed the branding and marketing functions for several top musical instrument and pro audio brands. He enjoys making strange noises, reading early Wittgenstein, and listening to George Jones—often all at once.

Connect with him at?www.dhirubhai.net/in/audiohawley