The 13-Step Vocal Production Workflow
It’s a bit of a scary task taking on a topic like this because there are so many different paths you can take when recording and mixing vocals. I’ll be assuming you’re familiar with your DAW, but now you want to start adding vocalists to your productions. I’ll be holding your trembling crybaby hand, all the way from the beginning of the recording process, to the end of the mixing process.
Don’t be discouraged if your first attempt at this doesn’t turn out as you had hoped. If you’re recording a quality vocalist, your main job is to record and process their vocals WITHOUT SCREWING THEM UP! Seriously, your first priority is to make sure you don’t accidentally destroy the vocals you recorded. Poor quality recordings and over-processing are the two biggest offenders when people end up with a low-quality results.
Sky Pilots is a project I'm apart of with my girlfriend Alyssa. She takes care of the songwriting/vocals, and I look after the production/engineering. I used the techniques in this guide to achieve the following vocals. If you like the track, you can download/stream "Falling Hard" here.
Recommended Shopping List
The following list of items is what I use, and is capable of providing professional quality results. After a lot of trial and error, I’ve narrowed down the seemingly limitless number of options available on the market to a list of my favorites. Feel free to swap in other plugins and devices at your discretion:
Time and Pith Correction Plugins
Return Track Plugins
1. Acoustic Treatment
Before you even touch a microphone or contact a vocalist, you need to be able to record in an acoustically treated environment. This doesn’t mean that you need to drop $50,000 renovating your house to achieve some sick fuck’s idea of “the perfect acoustical environment;” it simply means that you should put in the smallest bit of effort to make sure that your room doesn’t sound like a tin can.
You can easily treat your room with moving blankets which will cost you around $60, or DIY acoustic panels for around $400. For more information on this, as well as the best panel build I’ve found online, check out The EDM Artists Guide to Studio Gear.
2. Audio Interface
All you need is an audio interface with at least one XLR input. If your audio interface doesn’t have one XLR input, you’ve been trying to plug your microphone into a potato for the last 20 minutes. Go home, you’re drunk.
The Scarlett 2i2 is dirt cheap and will get the job done. All you’re using this for is converting an analog signal into a digital signal. Your computer doesn’t understand electrical signal, but it understands 1s and 0s. An audio interface performs this conversion.
Getting your hands on an awesome microphone isn’t hard or expensive. Go to Long and McQuade, rent a couple mics ($15+/mic) for your vocalist and pick whichever one fits their voice the best. You’ll also need a microphone stand and a pop filter.
Require 48V phantom power. Push the “48V” button on your audio interface or pre-amp to enable it.
Do not require 48V phantom power. However, some mics like the Shure SM7B require a ton of pre-amp gain.
Make sure your microphone is set up on a microphone stand and has the name of the microphone facing your vocalist. If your microphone is facing the wrong direction, you’ll be recording your room, not your vocalist. This is a simple, yet critical error.
Attach a pop filter to your microphone stand and position it 6” away from the microphone. The vocalist will sing through this pop filter into the microphone. A pop filter reduces sibilance and it also helps stop your vocalist from getting too close to the microphone.
The distance you position your vocalist from the microphone will completely change the character of the sound captured. Positioning them closer will provide a warm, intimate sound, while positioning them farther away will provide a thinner, far-off sound.
Something that’s caught my eye recently is the Slate Virtual Microphone System. It uses a condenser microphone, sonically-neutral pre-amp, and a digital processing suite that recreates the tone of classic microphones and preamps. For $1000, this can be an excellent choice if you plan to work with various different vocalists.
4. Recording: Levels
Now that you have your microphone plugged into your audio interface, you can start recording your vocalist. Your one and only job is to make sure that the audio running into your DAW doesn’t clip. Keep your track level at 0dB and make sure your audio is being captured at a level of -18dB to -12dB. This will ensure you’re taking full advantage of the dynamic range available to you, while still leaving a safe amount of headroom.
Record quiet parts with a higher input level and record louder parts with a lower input level.
Get an analog compressor and attenuate transients before running signal into your audio interface. Doing this creates a less dynamic signal, making it much easier to track vocals properly. DBX has a range of products that are around $200-300, and they get the job done. On the higher end of the spectrum, I’d recommend something like an EL8-X Distressor ($1600).
5. Recording: Vocal Takes
This one is pretty obvious, but you need to record your lead vocal first. Every other vocal take you record will be built off of this one, so it's important to get it down. Your vocalist will probably want this take playing when they record the rest of their vocals, and that's totally ok.
Lead Layers (Left/Right)
The idea here is that you record two extra versions of the lead vocal and pan them hard left and right in the mix. This will create an intense width and thickness that's not possible to achieve otherwise. Using a harmony plugin like Nectar won't work as well as recording the real thing. The subtle variances in pitch and time are what will provide this effect.
You've probably heard this a million times in different songs. At the end of a phrase, a vocalist will "double" the lead vocal by singing over it. Using a different vocalist to record doubles can have a nice effect because their voice will contrast the original vocalist's voice.
There are so many directions you can go with harmonies and they can be as simple, or as complicated as you want to make them. A simple harmony would include "ooo" and "ahhh" in the background, while more advanced harmonies would include different vocals, stacked to create chords that change throughout the song. Ask your vocalist to help you out with this. If they're experienced, they should have a good handle on creating harmonies.
If you don't know what ad libs are, listen to Gucci Gang by Lil Pump. In the background, there are ad libs throughout the whole track. All those one-shot vocal chants are ad libs. These are really popular in Hip-Hop and can also be found in other genres as well.
I like recording into Logic because of a feature called comp folders. Comp folders let you easily stitch together your favorite parts from different takes. This saves a tremendous amount of time. Although I produce in Ableton, I would never even think about running a vocal session in it. It just has no way of dealing with vocals the same way that Logic does.
Unless your running your audio through an analog gate (which the DBX 286s has), you’re going to have to manually apply fades to remove ambient room noise from your audio.
8. Gain Automation
There are two things you need to do now. You want to bring down the level of the breathes your vocalist is taking and you also want to attenuate any really loud peaks in your audio files. Putting in the time, and dialing in levels like this manually, will allow compressors you apply later on to perform better; the result will be a vocal with consistent levels.
9. Time Correction
Once I’ve comped my vocals in Logic, I’ll bounce the stems back into my Ableton project. At this point, I’ll warp them and snap the transients to the grid of my song. Make sure you listen to your audio while doing this because you can ruin the natural timing of the vocals. Sometimes timing that’s a little bit off sounds more organic and desirable.
Vocalign Pro tightens up the pitch and time of audio by matching it to a reference track. This will allow you to create lead layers that are perfectly in time with your lead vocal. The result is a super thick, wide lead vocal that appears to be one voice. You can use this for doubles and harmonies as well.
Revoice Pro is made by the same company as Vocalign Pro but allows you to process more audio at once in a program dedicated specifically to this purpose.
10. Pitch Correction
Even if your vocalist is great, you’ll probably need to tune their vocals a little bit. Melodyne is pretty much industry standard because of how easy it is to adjust individual notes and fine-tune small nuances in your vocalist’s performance. It comes with a rather expansive interface, so I’d recommend you watch a couple tutorials before using it.
I’m recommending AutoTune as well because I’ll actually use this in serial with Melodyne. Once I’ve cleaned up everything to my liking with Melodyne, I’ll apply AutoTune in very slight moderation to really tighten everything up. I find that AutoTune works great as an overall effect for a track, whereas Melodyne works better as a precision tool.
11. Cleanup Plugins
Quite honestly, your stock EQ will work fine. I usually low-cut around 100Hz for male voices and up to 200Hz for female voices. This removes some of the low-end mud that takes up space in the mix. If for some reason your stock EQ isn’t cutting it, Fab Filter’s Pro-Q 2 will be a nice upgrade.
To add a bit more air and presence to the vocal, I’ll apply a gradual high shelf and boost it around 5000Hz. This can help it cut through your mix; just be careful not to make it sound too “thin.”
A de-esser attenuates sibilant peaks in audio. It does this by responding to audio that peaks above the threshold of a band set to around 7-10kHz. I highly recommend the Fab Filter Pro-DS, but if you’d prefer to make you’re own for free, check out this guide on How to Fix Harsh Vocals.
Vocal Rider is one of those magical plugins that’s so simple, yet so useful. It rides the gain of audio being fed into it, pulling it towards a set level. This should help reduce dynamic range enough so that when you finally add a compressor, it doesn’t need to work very hard to give you optimal results.
12. Character Plugins
At this point, your vocal should already sound really good. If it doesn’t, something has gone wrong during the recording, or cleanup process. What you’re doing now is adding extra character to your vocals.
I like using two compressors on vocals. The first compressor will attenuate peaks, and the second compressor will manipulate the body and decay time of the vocals. I usually use Fab Filter’s Pro-C 2 to attenuate peaks because it’s extremely responsive. It will clamp down on transients as soon as they pass above the set threshold, and then release them quickly as well. The second compressor I use is usually one that is rich in character. Slate Digital’s FG-Stress is an emulation of Empirical Lab’s EL-8 Distressor. A little brighter than the analog gear, it still provides a wealth of character to your vocals.
I love applying a tape saturator to vocals. It’s a classic technique that really warms, and thickens them up. Slate Digital’s Virtual Tape Machine is what I’ve been using lately. The more you drive the input, the more saturation you apply.
I’ll actually apply a slight amount of limiting on my vocal buss. When multiple different vocal tracks sum together, they can sometimes create unwanted transients. Use this limiter to catch those peaks and tuck them back into the mix. You don’t want this limiter applying constant limiting; it’s meant to be used solely for catching transients.
I like using the Brainworx bx_limiter for this purpose. It allows you to apply limiting without makeup gain. This makes it easy to hear what it’s doing to your audio.
13. Return Track Plugins
I apply all effects to my vocals using return tracks. This makes sure that the integrity of the original vocals remain unaffected, while still allowing me to apply effects, and manipulate the wet signals independent of the dry signals.
I’ll use a short reverb with a decay time of less than 1s, and send a little bit of signal from all my tracks to it. This creates the perception that all of my song’s elements are living in the same space. It helps to make the song “believable.”
The second reverb I use is longer, usually with a decay time of 3 or more seconds (depending on how spacious I want the track to be). Elements that I want to appear further back in the mix will have more signal sent to this return, while elements that I want to appear forward in the mix will have less of their signal sent to this return.
Using a delay on vocals can be great if you want to give the vocal space, but don’t want it drowning in reverb. A short delay of 15-30ms can push the vocal forward and thicken it up, while timed delays (such as 1/8th notes, 1/4 notes, 1/2 notes, etc.), can set them into the mix without washing them out.
I apply parallel compression to my vocals, as well as other elements in my song that I want to add body to. This helps these elements stand out in the mix. The idea is that you apply heavy compression to the signal running through this compressor and mix it back in with the original signal.
Start with a ratio of 10:1, attack of 30ms, and a release of 200+ms. You’re looking for about 10dB of gain reduction here. Tweak these settings to your liking and to the material you’re working with.
If you’ve made it to the end of this article without your head exploding, congratulations! You now have the ability to create high-quality vocals all on your own. Refining this process so that it works for you is something that will take time. You can achieve vastly different sounds just from the pre-amp you decide to use. Each decision and piece of gear you use along the way contributes to the sound of the final product. The best way to find out what works for you is by experimenting, screwing things up and then finding solutions!