Golden Vocal Chain
You may not be privy to this (especially if you’re primarily a beat-maker), but the vocals are arguably the most important part of your track. Of course melodies and groovy drum rhythms can get stuck in your head, but it’s the words, emotions, and melodies that come from the vocalist that are really ear-bug worthy. That’s why in this blog post, I’m going to be sharing with you a standard vocal chain I’ve been using for a while now, and it’s been working great for me. I start with this chain every recording session I host and most of the time it works perfectly with my remote mix clients as well! Check it out!
- Gain Stage – Make sure the input level from your recorded vocal track is sitting at about -12 to -18 dB (dry, no effects or compression yet). This is to ensure that you’re recording with ample headroom. If the rest of the mix is overpowering the vocals – DO NOT turn up the vocals!!!! I can’t stress this enough, as it’s a quick route to leveling problems later on down the line. Instead, just turn down the instrumental track! Alex Tumay (Young Thug’s engineer) talks about this in a couple of his interviews in reference to recording over a two-track. You don’t want to have your input so high that you risk clipping on the way in. A common request from recording artists is that they want to hear themselves more in the headphones. Simply turn down the beat! You’ll be turning everything up on your mix bus, so don’t worry about putting out a quieter track. You’ll be adding more elements to the mix, like background vocals, adlibs, and maybe even some gun sounds. If you’re clipping your master fader, chances are you’re not monitoring the way the track will actually sound. This is super apparent when mixing kick. If the kick is clipping out the master fader, it’s doing some limiting/soft-clipping, depending on what DAW you’re using, so you’re not even hearing the kick the way it is actually sitting in the mix.
- High-Pass / Preamp – Depending on your hardware setup, you may have a hardware preamp, or you may not. If you don’t, consider using a digital preamp to add some color and tightness to the vocal signal right off the bat. This isn’t really necessary, especially if you have a solid mic, but it can help! I notice that when using the Neve 1073 emulation from Universal Audio, there is considerably less background noise in the dry audio signal, and a couple of useful utilities built in to the interface. I wouldn’t recommend EQing with a digital preamp, but I would highly recommend putting a high-pass filter to use! Somewhere between 80hz and 120hz is a great place to high-pass, as this will reduce unwanted boominess, and further tighten up the vocal. It’s also a good idea to use a gate at this point to get rid of any unwanted noise when the artist is not performing. You don’t really want to use a gate before you high-pass, otherwise some low rumbling sounds could possibly trigger the release of the gate and allow noise through.
- EQ – At this point in the chain, it’s a good idea to start shaping the vocal. Oftentimes, a 3dB bell dip is called for around the first harmonic (usually around 400-600hz). We don’t want this area to trigger our compression later on down the chain. I would NOT recommend low-passing, UNLESS you’re using a very cheap microphone that has some absolutely disgusting information coming through above 15khz. In fact, a slight boost (3.dB) to the high frequencies (high-shelf boost starting at around 2-4khz) can be a great way to get some instant clarity to the vocals.
- Compression – This is where a lot of people screw up. They think that crushing their vocals here will make them sound thick and full, when in reality, compression is volume control. There are a million ways you could go about compressing vocals, but I’m going to share with you the most effective way that I have found. We’re actually going to use 2 compressors! The first one should be a FET-style, medium to fast-attack compressor. My absolute favorite is the 1176 Rev E from Universal Audio. I will almost always have the attack speed set to around 5 (fast), and the input gain such that the compression meter reads about 3-5dB of compression at its peak. You want nearly every word to move the needle at least a little bit. This will ensure that you’re squeezing the peaks down, and evening everything out. The release knob will vary from song-to-song, but ideally, you want it between the 7 and 4 setting. You’ll have to use your ears, but a good rule of thumb is that the compression meter should fall with the words, and not stick around too long. On fast rap, you will want to use close to the fastest release possible. On something slower, like melodic singing, you’ll probably like the sound of a slower release better.
- Compression (continued) – Moving on to the second compressor! I have found that once the main peaks have been tamed with a fast-acting compressor, a slower, optical compressor sounds incredible, and does a great job of lasering in the vocals dynamically, bringing them to the front of the mix effortlessly. My go-to is the White 2A Levelling Amplifier from IK Multimedia. Trust me when I say I’ve tried DOZENS of 2A emulations, and this is by far my favorite. You could absolutely use one of the emulations from Universal Audio, or Waves, but they honestly just don’t do it right. With the 2A, the goal is to squeeze the vocals a little more. We’re not chopping off sharp peaks, and we’re not trying to distort the vocals at all. We’re aiming for 2-3dB of fairly consistent compression. More often than not, you’ll just have to turn the peak reduction knob to the left a little bit, and then make sure your gain is not out of wack. You want to exit the plugin with the same amount of gain that you entered with.
- Multiband Compression! – At this point, I often find it very effective to ride the first harmonic with a multiband compressor. Make sure it’s fitted in with the low-mids, and then use moderate attack and release settings to make sure it stays there! I use the Fabfilter Pro-MB, and pretty much just stick with the default attack and release settings. A similar process can be used for the high end of the vocals by using a de-esser. Don’t get things mixed up – a de-esser is just a multiband compressor! My go-to de-esser is the Pro-DS from Fabfilter, using the “Audition” function to find the most annoying frequency. Hint: it’s usually between 4-6khz. It’s up to you how wide you want the band, but in general, I let the band go up to around 10khz.
- Post-Compression EQ –After you’ve levelled out the vocals with compression, it’s time to really dial in how you want the vocals to fit into the mix. We’ve already done a bit of shaping pre-compression, but that’s not where you should be focusing on fitting the vocals in quite yet. It’s far easier and more effective to EQ after compression, because there won’t be as many peaks throughout the audible spectrum. Everything is locked in, and ready to be shaped in the context of the mix. If there are any glaring resonances, consider using an EQ bell cut with a tiny Q factor, and just dip out the resonance 10-12dB as necessary. Don’t go all willy-nilly just cutting everywhere. I’ll usually allow myself 2 cuts maximum, otherwise there could be other issues earlier in the chain that are causing such resonances. When it comes to boosting, I find it most pleasing to use something with color. My go-to is the Pultec EQ Pro from Universal Audio. This is a combination of the EQP1A and the MEQ-5. In order to get the high-end of the vocals to stick out a little more (or a lot more), it’s really effective to use a wide-band, and boost around 5khz. Before you do that though, consider using the top module, the MEQ-5 to tame the mids. This is the most audible part of the audio spectrum, and it’s not a rare occasion that I find myself taming 1.5khz before moving on to the highs. If you find that, after boosting the highs, you have TOO much air to the vocals, it’s a good idea to nip that right in the bud. Set the attenuation to 20khz (sometimes 10khz works better, but I find that it does too much), and use the attenuation knob to dull the vocals a bit, as needed. This will allow you to tuck the vocals in, and fit them in with the hi-hats appropriately.
- Finish Up The Dry Vocals – At this point, you’ve probably been locked in to mixing these vocals for a while, maybe 15-20 minutes. Take a walk. Go grab some water, some coffee, take a smoke break, whatever you need. Just go out of the room for a couple minutes, and come back in with refreshed ears. Trust me. When you come back, it’s time to make sure the vocals are absolutely rock-solid. You may want to add another de-esser, you may want to tuck the high end in a little more, you may want to dip out a little more 400hz… You’ll be much more accurate in your decisions if you take a quick break. I find myself taming the high end a little more when I come back with either the Pro-DS or the C1-sc from Waves (check out the Mary Vox preset!).
- Effects – This is probably what you’ve been waiting for… Effects on vocals!!! Delays, reverbs, automated distortion, you name it – now is the time to add these effects. My favorite way of applying reverb is as a send. Using either an Aux bus, or an effect rack in Ableton Live, I’ll apply between -15dB to -20dB of 100% wet reverb. This maintains the perfect vocal we just mixed, and allows you to process the wet effects completely separately. This is HUGE! Don’t get lazy with your sends! The same goes for delays. Set up a bunch of different delay sends (1/8th note delays, 1/4 note delays, 1/2 note delays) and use them by automating their faders. You can even save an effect rack in Ableton so you don’t have to do this every mix, you can just call up the rack, and put it to use right away with everything pretty much set up from the get-go. When applying distortion, funny enough, it’s usually better to apply this directly to the vocal signal, instead of sending it to an Aux bus. I find it much more effective to change the signal, and put the listener in a different environment for a brief moment before bringing them back to the original vocal that they know. Automation is key.
- Grouping / Bussing – I’m not huge on bus processing for vocals. The only time I use it will be to EQ a resonance out of a crappy mic that builds up over the course of 4-6 simultaneously playing vocals. A lot of time, 4.5khz will build up on Rode mics, and other cheap brands. You won’t notice it until you have everything pretty much done. You might even try to start cutting it out of each track individually before you realize… wait! These are all the same resonances that I’m cutting out! Save yourself the time, and EQ on the bus. BE CAREFUL because this will apply the changes to every vocal in the bus. So only do this if you have to. The other instance in which I will use a full, “All Vocals” bus, is for leveling! Say you want to turn down all the vocals about 2dB. It’s much easier to just turn down the bus, and this ensures you’re turning everything down congruently.
Do you feel like a pro now? Do you think I forgot about AutoTune? I didn’t. It just didn’t deserve it’s own section. Put it at the beginning of the chain, right after your high-pass EQ/preamp. Use Autotune 5 or Evo (or a combination of both) and don’t let anyone tell you otherwise. You’re welcome.
Peace! Happy mixing.