How to Record Pop Vocals, Pt II — Editing

JMC Academy alumnus, Peter Holtz, uses three tools to complete the triumvirate of pop vocal compiling, timing and tuning.
Column: Peter Holz
Last issue, we went over what it took to get the best vocal takes for Peking Duk’s multi-platinum singles. In this part, we deal with the raw materials you’re left with once everybody else has walked out the studio door.
During the tracking stage of a densely arranged pop song, it’s fairly common to frantically fly all over the place without finalising the comps or separating out all the layers for your stacks/doubles. Sometimes you haven’t even put a comp together at all.
You do need to be confident you’ve got everything before the artist leaves. If you’re tracking against a guide (usually from a song pitched by a publisher), always cross-check as you go to ensure you’ve nailed all the right parts, words and melodies. 
I usually break the editing part into three sections:
1. Compiling
2. Time adjusting/aligning
3. Tuning
It is not uncommon for this entire editing process to take two to three times longer than it did to actually record the vocals. The key points to remember while editing is that on a modern pop vocal it is simply assumed the pitch should be ‘in’; the delivery and rhythm/timing are the most important.

STEP 1: COMPILING
[fig-1] It may seem obvious, but compiling is the process of flicking between take playlists and putting together the best possible version. There is no set way to do this as it is extremely subjective. The main thing to watch out for is unnatural and unintentional overlaps. These often occur when you’ve recorded sections separately without the singer singing in and out of each section. It usually manifests as them breathing in to sing the next line as they are still finishing the previous line. I circumvent this by getting the singer to sing into the next section, if only for a bar or so, when looping takes.
Precise edits also help maintain the illusion of a consecutive take. The illusion will start to disappear if the listener is distracted by an unnatural pause, a cut off breath, or a consonant at the end of a word that’s been trimmed.
It is possible to perform edits that look like they shouldn’t work. Sometimes a sound taken from elsewhere in the song — commonly a ‘t’, ‘s’ or a breath — can paste in seamlessly. Just close your eyes or turn off the screen. If it sounds okay, then it is okay. When you’re staring at a bunch of heavily edited clips, you can often be fooled into thinking you can hear something that’s not actually there.
STEP 2: TIME ADJUSTING/ALIGNING
Timing is paramount when laying vocals over tightened pop music. I’m not saying you need to quantise the vocals, but the groove and feel of the vocal need to be as ‘in’ as possible. It’s not just the drums and bass that need to be in the ‘pocket’.
If your singer is perfect and hit every single part with the exact right groove then skip this section. Aim to get this sorted out as much as possible whilst recording, but you can always take it a step further with some editing. 
If the take is ‘pretty close’ then I will always try and perfect it. One caveat: it is possible for badly executed time editing to make things sound extremely odd. Like compiling, the nuances of timing are subjective. It’s not uncommon for me to spend 15 minutes working on the lead vocal timing over four bars.
I do all of my timing adjustments by performing micro stretches with the time tool in Melodyne. The approach I use goes like this:
1. Get the lead vocal timing exactly as you want it.

[fig-2] Print that and use it as your ‘guide’ in Vocalign.
3. Align all doubles, layers and harmonies to the lead vocal.
4. Bring all of those tracks back into Melodyne and finesse the timing even more. 

[fig-3] There is a little bit of a tonal difference between Melodyne and Autotune, but the reason I use Melodyne in this stage is because it allows you to see and edit multiple tracks at once. Grabbing a stack of 16 vocals and adjusting them all together was a game changer for me. [fig-4] Another benefit is having your lead vocal as greyed out ‘guide’ blobs underneath any doubles or harmonies you are finessing.

STEP 3: TUNING
If the word ‘tuning’ makes you baulk and instantly want to chastise someone about ‘getting it right during the take!’, then feel free to ignore this whole article; it probably doesn’t apply to the style of music you are making.

The countless hours I’ve spent tuning vocals in Melodyne has made it almost automatic. [fig-5] I primarily only use the Pitch and Pitch Drift tools, which means I’m rarely changing the actual curve the artist sang, just offsetting it. As soon as you adjust or flatten the curve, you will hear it.
Occasionally, I’ll use the Pitch Modulation tool to limit vibrato. I will also use it to lock a quick passing note the singer may have pitch bent over.

[fig-6] After the notes have been massaged with Melodyne I will usually feed that signal into Antares Autotune to give it a final squeeze and impart the tone Autotune seems to give. It’s something we’re all used to now! Importantly, I’m never using Autotune to correct or move any notes, it’s just swimming over the top. Also, make sure to keep each stage on a different playlist so you can always ‘go back’ if you need to. It’s basic house keeping.
HORSES FOR COURSES
Most instrumental elements in a modern pop song are more or less perfectly in tune. For this reason, even the best singers will still benefit from some pitch correction.
However, it’s still down to the actual song. Once, I recorded an extremely talented singer on two songs in the same day. One track was a modern electronic pop production which I rigorously layered, edited and tuned. The other was a more laid back soul tune with an acoustic piano as the main element. There was no need to tune anything on this second track as the natural tuning variations of the acoustic piano allowed the vocal performance to have more room to move in terms of pitch. It was a wider lane to move around in.  

AUTOTUNE PRO TIPS
1. I will often ‘hard tune’ some of the doubles and layers in Autotune and slightly reduce or increase the scale offset amount for different layers. This gives them an incredible tightness with some width and depth due to the slight pitch offsets. 

2. [fig-7] On doubles and layers, I also use the ‘create vibrato’ section in Autotune at a very slow rate to make sure it never holds a constant pitch offset. This acts like a slow phaser/flanger that’s ever so slowly modulating changes in pitch.

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

Read More
Part 1
Part 3