When mixing a 2 track beat and vocals, be sure your beat has enough headroom before adding any processing to it - ideally using clip-gain to reduce its level. Then use multi-band expansion to increase its dynamic range, and manipulate its side and mid image.
The chapters in this video are a chain, so we’ll look at each step one by one and build on our processing with each demonstration.
The first thing I want to do is reduce the gain of the instrumental - I’ll use clip gain for this since it occurs before any processing that we’ll introduce, whereas the volume fader will occur after any and all processing. By lowering the gain, we give ourselves more headroom.
Before we start processing, let’s just get acquainted with the track we’re working on today, and solo the vocal and instrumental.
This step is more or less optional, but I find it helps the overall sound - first I export the reduced gain instrumental and import it into Izotope Rx. Then I perform subtle de-clipping which is going to be especially valuable if you pulled the track from YouTube or Soundcloud.
I don’t want to repair every clip, just those associated with the peaks. While we’re here, we might as well fix the phase rotation, by clicking suggest, and then rendering the changes. This will help reduce unwanted clipping when we start to process the instrumental.
Let’s do a null test between the original instrumental and the de-clipped one to hear the differences, and notice the occasion clipping in the left or right channel.
Back in our session, let’s add our first processor to the 2 track beat - which should be a multi-band dynamics processor. I use the Pro MB and create 5 bands, all of which have their gains reduced, but include expansion with a positive range.
By lowering the gain and then introducing expansion, we increase the dynamic range while keeping the overall level of the track about the same. For the lows, I used a longer release and tried to time it to the track - for the highs I used a quicker attack and release to retain detail.
Lastly, I increase the amount of the effect slightly with the mix slider.
Let’s listen and notice how this keeps the frequency response of the track more or less the same, but it increases the dynamic range a good amount.
Next, I want to clear some space for the vocal by equalizing the beat - with a Mid-Side EQ, I’m going to cut away frequencies that will interfere with the vocal, but only do so on the side image. I’ll attenuate roughly 2kHz on the side, as well as 250Hz.
To find the frequencies that interfere with your vocal, listen intently when centering your band, but know that 200-300Hz and 2-5kHz are essential ranges.
Let’s listen and notice how this gives the vocal more room.
You may be wondering why I only equalized the side image in the last chapter - I still need to attenuate some frequencies on the mids, but I don’t want these frequencies to constantly be attenuated since they’re pretty important for the beat. So, I’ll use dynamic bands on the mids.
Additionally, I’ll use the vocal as the external side-chain, so that the vocal’s signal triggers the attenuation of these frequencies, meaning they’re only dipped when the vocalist is performing
Let’s take a listen.
Part of getting the vocal to sit in the beat includes slightly widening the beat - to do this, I like to use a parallel send, and with the free plugin MSED, I’ll mute the Mid image, in turn soloing the side image. Then I’ll insert a transient expander.
With the transient expander, I’ll amplify the transients, in turn dynamically widening the beat, and shifting some of its energy away from the mids.
This both sounds great and gives the vocal more room in the middle, so let’s take a listen.
I know it’s strange that I’m introducing clipping to the beat when I tried to reduce it earlier, but there’s a big difference between clipping introduced from bad encoding on the internet, and clipping introduced with a processor that’s designed to make it sound good.
With the PSP saturator, I’ll select hard clip, center low frequency saturation over my 808 to give it some weight, and do something similar with the highs to accent the transients.
Additionally, I’ll find and engage the FAT function, which in this plugin is oversampling. This will keep the highs clean and free from aliasing and phase distortion.
It’s taken a while to get to the vocal, but let’s start with compression followed by de-essing. For the compression, I want to capture the vocal quickly, so I’ll use a quick attack, and a 50ms release to retain the detail of the track. I also enabled lookahead.
This will help the compressor anticipate the incoming signal, allowing more to be captured by it. Then I’ll turn on auto make-up gain.
This vocal was already somewhat compressed, so I only needed 4 dB of compression, but use a little more, up to 7dB if needed.
For the de-esser, I’ll attenuate about 4dB as well, using a split band setting to attenuate only the selected frequencies and the set that ranges between 6-14kHz.
Let’s take a listen to how this dynamically controls the vocal and brings it forward in the mix.
Let’s start shaping our vocal with EQ, starting with a high pass to cut out some of the unneeded lows. We’ll also attenuate roughly 200-700Hz only by a couple of dB since these frequencies often cover up or mask the higher, more clarifying frequencies.
With that said, I’ll amplify 2-5kHz to add more vocal clarity, while dipping a little bit of sibilance frequencies. Lastly, I’ll add a small amount of air from 12-30kHz.
Let’s take a listen to how this improves the vocal and helps it stick out when played with the 2-track.
There's something missing from this vocal, and when that’s the case I often find saturation helps a lot. I’ll add Arturia’s new tube saturator and with the first setting enabled, slowly introduce saturation with the drive and bias dial - while ensuring the EQ is off in the advanced section.
The air setting helps add some clarity to the vocal,and the harmonics this plugin introduces both fill the vocal and help separate its frequencies from the instrumental's frequencies.
Last up, I want to add high frequencies to the vocal, but I don’t want them to be harsh or unpleasant - I’ll first set up a bus, and on the aux track for that bus, introduce a linear phase EQ that isolates the highs. Then, I’ll compress.
I’ll solo just this aux track while I’m compressing so that I can hear what compressor settings result in the smoothest and most enjoyable sounding high frequencies. I’ll typically use a longer release, softer knee, clean compression, and turn off auto-make-up gain.
Then, I blend the processed highs with the original track via the fader.
Let’s take a listen to what these highs add - then let’s take a listen to how the vocal and beat sounded when we started, and how they sound after our processing.