How to Record Pop Vocals, Pt I
JMC Academy alumnus, Peter Holz walks us through how he records multi-layered vocals for Peking Duk’s multi-platinum singles
Column: Peter Holz
It was Christmas, 2015, and I got a call asking if I could cut a vocal with singer Elliphant for an upcoming Peking Duk song. As I have no actual life and never take holidays, I was there with jingle bells on.
This wasn’t just an in-and-out engineering session. It was a writing session where the lyrics and melody (or topline) were being written over a pre-existing track. 12 hours later we had a completed, multi-layered punchy dance/pop song with close to 100 vocal tracks.
Over the course of the day and a bottle of rum, the four of us — Adam and Reuben (the Duks), Elli and myself — formed a great working relationship. A few months later they sent me some more vocals to edit and mix. That became the song Stranger, which went on to win Song of the Year at the 2017 ARIA Awards. We kept working together and managed to roll out three back-to-back multi-platinum singles. Who needs presents with gifts like that!
IT’S ALL AN ILLUSION
When we’re recording and mixing pop vocals, the aim isn’t to make things sound real; we’re creating an illusion, a powerful fantasy. If it sounds right, it is right! With most electronic-based music, you’re trying to combine vocals, which are fundamentally imperfect, with music that is more or less ‘perfect’ in terms of timing and pitch. While we’re not restricting ourselves to real in the sense of ‘untouched authentic’, it is important to remember that people will connect with something that feels real to them. Even if it is layers of stacked vocals with autotune, formant shifting and vocoders.
There are techniques to make a less desirable vocal sound ‘good’ to a listener, but the ability to make a performance connect to the core emotion rests solely with the vocalist, not the editing/mixing.
I’ve decided to break down my specific workflow into the three staples of recording, editing and mixing. I’m not saying that these are the right ways to do things (there is no right way!), these are merely techniques I’ve developed, learned, copied and put into practice many times, then distilled down to what works for me. This issue, let’s get stuck into recording.
PREP TO IMPRESS
Being prepared is everything. It doesn’t matter how great your gear/space is or what you’ve done before — you need to be ready.
Being ready means:
Having the song/session laid out: In your head and on the computer. Know the structure, the key, the tempo and the majority of the words and general meaning.
Have the studio ready: Microphone set up, headphones mixed and checked, lyrics printed and on a music stand, phone/computer power near vocal area, and bottled water ready. Candles, lights, snacks, tea/coffee… you can never be too prepared.
Important: You must always know the Wi-Fi password.
Most importantly, you need confidence. You won’t always have the luxury of arriving an hour or more in advance to setup. You should be able to walk into a studio at the same time as the artist completely unprepared and still get it all up and running smoothly, calmly and quickly.
Your personality and vibe is vital. People have to want to hang out with you in a confined space for hours on end getting deep into their thoughts and emotions. Leave your ego and attitude at the door and embrace mutual respect. In my experience, success and talent is proportional to humility and work ethic.
DON’T SPIKE THE MIC
I almost always use a condenser mic. Sometimes a Shure SM7 dynamic can do the job but it can get a little angry sounding when really compressed in the final mix. Don’t get too picky about mics unless you are in a situation where you have the time to compare a few. Throw up what you have, or whatever is closest and just get on with it.
The key thing to watch is making sure the singer doesn’t get too close to the mic. If a singer eats the mic, it results in massive sibilance and an implied ‘overload’ or compression. An experienced singer will know how to dynamically work the mic, and if the part is gentle it might make sense to cosy up to it (see Jeff Buckley’s Hallelujah). How to avoid it? Talk about it with them, watch them, and listen to what’s coming through the mic. Crucially, if the singer can’t hear themselves in the headphone mix their natural reaction will be to move in closer. Turning them up can help.
Sometimes singers will be able to overload the amp circuitry in a condenser mic, and you’ll need to instantiate the pad. If I’m in a hurry and on the fly I will default to always having the pad on.
I personally never want to record vocals without an Empirical Labs Distressor. It’s hands down my favourite vocal tracking compressor. It gets in and out quickly and just does its job. You won’t hear it change the tone, even with loads of gain reduction. Most importantly, it’s quick and easy to adjust on the fly during a recording and won’t put clicks or crackle into the audio as it’s being adjusted. Sometimes you need to ‘learn’ the part and adjust it during takes. This can also be achieved by riding the output of the preamp, usually the fader.
In a perfect world I would follow the Distressor with something tubey and slow, ‘swimming’ on top of the signal and never moving against the vocals, like a Tube-tech CL1B, or Teletronix LA-2A.
The preamp is less important to me. I would assume it’s not a bus-powered small interface, though you can make that work. In a perfect world I would always choose some 10xx flavour of Neve. Transformers help ‘slow down’ the close miked vocal. The beauty of a unit like that is having a built-in EQ before your limiters, the order I feel works best. I would only ever EQ a vocal a little on the way in; commonly a high pass filter combined with a low frequency push, and a shelf to open up the high frequencies. Maag’s ‘Air Band’ is perfect for lifting the top end.
HEADPHONE MIXES THAT STICK
If you have the luxury of having the artist use their own remote headphone mixing system — like Hearback, etc — definitely go for that. You’ll need to spend time tailoring specific DAW routing, but the workflow is very beneficial. Sometimes singers just want you to control their mix. In that case, you can simply send them a stereo cue that they can control with the mixer’s master level.
I have a custom-built headphone mixer made from JLM components that allows the artist to adjust:
Vocal effects (stereo)
There is some debate that open/semi-open headphones allow the singer to feel more ‘natural’ as they’re also hearing themselves acoustically. The downside is that it’s going to spill into the mic more than when using closed headphones. It’s extremely common for the singer to have one ear off. In this scenario, you can minimise spill by routing the mix only to the side they’re wearing.
If the vocalist sings better with speakers blaring and just going for it in the room, then you do that… it’s just part of the deal. It’s rare for multi-layered pop music, but sometimes rough writing demos are cut like this and they are the ones you’ll need to use!
Talkback is a big deal. I like to have it on ‘auto’ where it’s always on when stopped and switches off when in play/record, with the ability to override it if I need to talk or sing whilst playing and recording. To do this, I route a talkback mic into the DAW and use Sound Radix’s Muteomatic (it’s free). You do need to let everyone know the talkback is always open. I’ve had a few occasions where someone in the room has forgotten…
DEAD ROOMS ARE BEST
In my view, most vocals should be recorded in a nearly completely dead space. This doesn’t mean a small space, just a dead space. I will commonly build a little ‘hut’ with gobos and rugs in the middle of a big room. Any setup where the sound of the room doesn’t get back into the mic, too much. Even a small amount of room ambience can hinder a recording, as once the vocal is compressed, equalised and up loud in a mix, that ambience can ruin the illusion.
Music stands should also be deadened; you’ll be surprised how much they can ring next to a loud singer. I like to clip a heavy pillow case onto it.
SHORTCUTS TO SUCCESS
During a tracking session things need to happen on the computer quickly for two reasons. Firstly, you don’t have time. Secondly, everyone loses faith in you if you fumble around, and you won’t get called back again. There’s nothing more valuable to your career than a good recommendation, and fumbling around will kill any chance of that.
I use a vocal tracking template that is fairly simple but large. The basic premise I use is that I will record different parts of the song on different tracks (usually in loop record and then comp together from the playlists). This means my template has over 100 preloaded audio tracks that are all labelled with common parts of a pop vocal arrangement.
A common workflow would be to get the artist to do a few warm up takes through the song. Sometimes you get the perfect take this first time, or at least a line here or there. Once warmed up, move onto doing the sections individually.
Occasionally, multiple takes recorded whilst looping can be used to create doubles. Other times you’ll need to comp a final take together, then get the artist to sing against that so the timing matches as closely as possible. A helpful tip is putting the ‘guide’ in just one ear.
I also have a simple busing arrangement that allows an artist to quickly and easily use the multi-channel headphone mixer. Just be sure you have the pan positions the right way around, and the microphone levels aren’t changing for the artist — it should be the bedrock of the headphone mix. You also need to be sure you’re hearing the same ‘backing’ send as the artist.
There’s a little golden rule when working a DAW in front of artists, producers, managers and A+R: When things happen quickly on the screen without your hand moving around on a mouse/trackball, people think you know what you are doing. Know thy shortcuts!
Well, that’s the first part wrapped. Stay tuned for more on the editing and mixing in the coming issues.
PETER’S DREAM CHAIN
Microphone: Neumann U47, U67, U87 or TLM 170; Telefunken ELA M 251; or Shure SM7, SM58.
Preamp: Neve 1073
EQ: Maag EQ4
Compressor: ELI Distressor
Limiter: Tube-tech CL1B, Teletronix LA-2A, Fairchild 660.
Disclaimer: In my own studio I use a relatively affordable Aston Spirit condenser or an SM7 straight into a UAD Apollo MkII interface, followed by UAD plug-in versions of my Dream Chain. Even when I work from other studios that have the actual hardware units I will still bring and use my Apollo. That way, I know exactly what I’m going to get and can recreate it in the future if needed. It also allows me to control preamp gain, EQ and compressor settings while hardly moving.
Bear in mind, I spent many years recording and mixing exclusively though analogue gear before moving fully in-the-box. It’s important to understand how analogue gear works so you can implement those ideas in the DAW. Most importantly, gain structure.
If you’re looking for a great way to automate your talkback workflow, download Sound Radix’s Muteomatic.
Apply for JMC Academy’s Audio Engineering and Sound Production, or Masters of Creative Industries courses, to get qualified with hands-on experience, study abroad options and internship opportunities. Intakes in February, June and September. Check the courses out online at jmc.academy/audioat