In this post, I aim to offer some solution to one of the more challenging parts of the mixing process. Though I am going to spend the next short while attempting to sound like an extremely educated authority on the matter, don't be fooled, I in fact know very little and should not be trusted. One bit of value I would like to be taken away by anyone who reads this, would be to play with the tools you have. I'm definitely the first to fall in to the pit of shopping new gear and software for hours, searching for that 'quicker fixer', or that next big sound. Truth is, if you've got Logic, you perhaps have more than you think (especially V9+), and when paired with the extra stuff that i know you already went and bought because you got that email saying ...last chance grey Tuesday super sale... 99% off for the next 2 seconds!. Enough chat, let's start looking at what we got.
What I am using for this will be;
- Logic Pro x
-Some rubbish cheap second hand monitor speakers with no balance
-AKG openback headphones
-Waves REQ 6 + Aphex Vintage Vocals plugins
While I have little knowledge available in my ridiculously porous brain, due to information not passing into it, I do, like you, have access to the internet. I will be relying on this to translate myself better, as everything i have learnt is from pressing buttons and wiggling dials, and I will probably need to use slightly more agreed on terminology like 'threshold' instead of 'squishy squasher'. My biggest takeaway from mixing over the past few years, is there is no one rule that will apply to everything, and this is especially true in mixing vocals. The applied processes for female vocals, will be different for male vocals. The frequency cut required for a female rock vocal, will be different from the cut required on a female jazz vocal. The compression for one female rock vocal would distort another female rock vocal... etc. You get the point. LISTEN TO THE MIX. That's the only way to truly understand what you will need to do to make the vocals sit.
For this discussion, I will be using a track by a group I've been working on recently. I've chosen to talk about this track, by this group, for a few good reasons;
- 2 female vocalists
- Fiddle is present in the mix
- The recording was done at one of the vocalists homes, using microphones you will most definitely have access to
- The musicians are really talented, which makes me look better by default
I will be supporting this post with audio and visual to better communicate differences. Firstly, when receiving the recording, i listen to the entire piece, from start to end, in order to understand transitions in the mood and energy. This is something I will probably do 5 - 10 times before working on a mix, but ultimately it's up to you. You really want to be somewhere between 'I know where the chorus is' to 'this is my new favorite song and I know all the words'. This will help massively, as before working on an instrument or vocal, it is good to know it's journey in the music, so as to not apply an effect early on, perhaps when the energy of the song is low, and then find it heavily compresses, or distorts the signal, as the energy of the track builds. Once I have this sort of understanding, my first step is to create bus sends.
What is a bus send? I dunno, the internet describes them as... 'A bus is basically a path in which you can route one or more audio signals to a particular destination. ... In Studio applications, busses can be used to group signals together for recording when there are too many channels of audio for them all to be sent to your multitracker/interface/soundcard.' (pmtonline.co.uk)
That's pretty much what I would have said too. Bus sends are helpful to spread the processing load across your pcu. It is much lower consumption to have 1 vocal track with 3 bus sends (totalling 4 auxes playing the same audio signal), than having 4 individual audio tracks, processing 4 seperate pieces of audio data (even though the audio data would be the same information). This therefore leads to my 2 lead vocal channels looking something like this;
For this first part, we will ignore the information we have at the top of these images, and only focus on the lower half. You will see there are multiple bus sends active on both channels. Don't be intimidated, it's probably part of musical law somewhere not to use this excessive amount, but for me, it sounded right. It is also important to add that this is a screenshot taken from the final mix, so many of these bus sends would have been added as the rest of the instrumentation was being mixed. I usually start with 4 bus sends when working on vocals, and typically these bus sends would contain these plugins;
-space design (stereo spreader)
These would typically be dealt with on seperate bus sends, offering you more intricate detailing in the sound when it comes to the reverb, or the stereo representation of the signal. Once these are set up, my next step would be to roll them all off to 0, and only have the signal coming through of the main audio track containing the audio file. Everything that is applied to this track, will then be reflected in the signal coming through the bus sends, which again saves a lot of processing power, versus running seperate eq's and compression on seperate tracks. It also makes it the logical place to begin processing the signal. Scrolling back up to the top of the screenshot, you can see what processes are being run on this audio track. You will see the following plugins (all of which come as part of logic pro, and any other professional DAW software);
- Channel EQ
- Adaptive Limiter
- Channel EQ (again).
The order you have these plugins is important, but not something I will go into great detail about here. The reason they find themselves in this mix is because for me, it gave the desired sound.
Truth is, i was lazy here. The style of the song is very gentle, therefore, I wanted to reflect this in the compression of the vocals. The compression applied is minimal, but nonetheless applied. The threshold is quite low, meaning any time the vocal is coming through, compression is being applied. There is also corresponding side chains being used on both compressors. What a side chain does, is pass another signal through the compressor, without passing any audio out.
Often I will use this technique on any signals
that will be fighting for the same seat in the
I find it to work particularly nicely with two lead vocal tracks. When each track sings, the other will reflect the energy of its counterpart. By running this process, you will find that your lead vocal tracks, though seperate in signal, will glue together much better, as each is creating space for the other. Side chaining is something that I really strongly recommend playing around with, and not just when working with vocals.