This article is discussing a very common situation for all songwriters, producers, engineers… we already have a finished backing track that is mixed and mastered, and need to add some vocals to it.
I’ve managed a team of producers and writers for many years and kept recommending to them to do stems of all of their backing tracks. Just Drums, Bass, rest of instruments, all without a brickwall limiter – that makes it so much easier to add a vocal for a songwriter. But of course everybody kept slamming the 2-bus, and when vocals where added, more destruction occurred… such a standard thing in the songwriting world, and nobody is doing the right thing.
Back to solving the issue… I’ll guide you through it step by step, and if you’re a Logic Pro X-user, you can download a free template that is designed to do exactly what we’re discussing here – use the link at the end of this post – the free template is only available for the next 10 days.
1. THE BACKING TRACK
Producers who create backing-tracks want them to sound as big as possible – most of the time they are fully mixed, mastered and level-maximized, and sound really full and “loud”. When you import them into your DAW software, you can see in the waveform that they are coming in at full level, maxing out the meters.
We need to remember that the producer, when she created the backing track, did not necessarily care about adding vocals – the producer just wanted the best sound for the track.
In other words, the vocal has a difficult time to cut through the full backing track – this is especially true if you’re a songwriter trying to create a vocal hook on a full blown EDM-track which was originally intended as an instrumental club-track.
Because we need some room for the vocals, we start by turning the backing-track down by – 15 dB using a Gainer-plugin in the first insert-slot of the channel. No worries – we’ll “claim” those 15 db back at the end of the mix chain, but for now we’re lowering it by 15 dB to get some “headroom” to start with (more about “headroom” and gain staging in THIS ARTICLE.
2. THE VOCALS
I wrote another detailed article exclusively about creating a signal-chain for your vocals, what to avoid while recording etc., so check that out for more details – all of it applies here. Whats worth noting here is that we are creating an aux-send from the vocal that does something specific to the backing-track which we’re discussing next…
3. MULTIBAND SIDECHAINING
You might have heard of the term “multiband-compressor”, and some DAWs come with one. A multiband-compressor essentially splits the audio into various frequency bands (like lows, mids, highs) which can be separately compressed before they get mixed back together. That way, you can for example only compress the low-end, and leave the mids and highs untouched.
The technique we’re using here goes even further – we only want to compress the frequencies the vocals are using (to make space for the vocals in the mids), and only WHEN the vocal-signal is actually IN the mix.
The vocals have most of their energy roughly in a range between 400 – 4000 kHz, and we’re creating a mid-band that covers these frequencies.
There are dedicated plug-ins who can do this job, like Waves C6 Sidechain, but we can get the same effect by sending the audio to three busses that we split out using simple filters (low, mid, high).
In Logic Pro (and all other DAWs), that it easy.
a) First, we send the backing-track audio to a mix-bus.
b) Next we create a track in the arrangement for that bus. (so we can duplicate it to get a total of three mix-busses)
c) Using “New Track With Duplicate Settings”, we create a total of three destinations for the SAME bus.
d) The filters in Logic Pro’s Channel EQ on each of these 3 bus-tracks are used to split the audio into three bands.
We need to make sure the crossover points of the EQ still lead to a linear frequency response. We’re testing this with an analyser on the master bus, while sending a frequency sweep through the “multiband”-bus.
e) Inserting a compressor on the mid-band.
And yes, the compressor on the mid-band is sidechained to our lead vocal.
f) The compressor settings need to ensure a linear frequency response when the compressor is inactive.
HOWEVER, whenever the compressor receives a sidechain signal via the lead vocal, it dynamically carves out space in the mids for the vocal.
This is what the frequency-response looks like when the vocal is active.
Just to make clear once again what this is for… we are sending the BACKING TRACK through this multi-bus, and whenever the vocals are in the mix, the level of the BACKING TRACK in the mids gets lowered depending on the level of the vocal – that’s why the compressor on the mid-band is sidechained to the lead vocal!
4. REVERB
The reverbs and delays of the vocals – which we’re adding in the mix – is also being sent through the multiband-bus. Which means even the reverb-signal carves out a space for the vocals in the mids whenever they need it.
5. ON THE MASTER BUS
Time to add our the 15 db of gain back by which we lowered the backing track – it will end up sounding almost the same as before. All we’ve done was lowering the backing track by 15dB, carving out space in the mids for the vocals, then adding back the 15 db and a limiter on the master-bus to catch occasional peaks.
I’ve created a free template for Logic Pro X and ProTools that applies all of the above – get the free download and use it whenever you need to add vocals to a backing track…
And if you found this post helpful, check out my bestselling book which will guide you through the entire mix methodology from DAW preparation to mix delivery, the eBook YOUR MIX SUCKS. It’s currently still available at introduction price.
