When mixing vocals be sure to mix in the context of a full mix and not solo the vocal when mixing - additionally, use some subtractive eq, compression, and saturation in that order. Lastly, utilize temporal processing like delay and reverb after all of the other effects.
If you recorded the vocal via an interface, odds are you have a good pre but it sounds a little too clean for its own good. If so, emulate a mic pre to add some character to the recording - this will add some of the harmonic distortion you’re missing.
There are emulations of just about any popular pre from the past from Neve to API, so take your pick.
Before we go any further, one big mistake made when mixing vocals is to solo it while mixing. Using the solo function is great if you’re trying to find very specific problem frequencies; however, you need to mix in the context of a full mix.
The reason being all of the signals will interact with one another, in turn affecting the vocal in a way you wouldn’t notice if it was soloed.
If you have multiple vocal tracks it’s important to set up a vocal bus - to do this, change the outputs of your vocals to a new bus. This bus will let you collectively process all of these tracks before the collected signals head to output.
So we compress the groups collectively or add other processing as we see fit.
Next, I’ll attenuate aspects of the lead vocal’s track with an EQ to get rid of frequencies I don’t want - best to do this to the individual tracks first instead of the bus we just created. Let’s do this for each vocal track until our vocals are balanced.
Then if an overall subtractive EQ is needed, we can use one on the vocal bus.
On the lead, let’s insert a de-esser to attenuate sibilance and create a more balanced sound. It’s better to attenuate this range with a de-esser as opposed to an EQ since a de-esser will attenuate the signal dynamically or only when the sibilance is loud enough.
Although each vocal is different, try to avoid more than 4 to 5dB of attenuation with the de-esser.
Next we’re going to use 2 processors back to back on either the lead or the vocal bus depending on what you need. I’ll use downward compression with a quick attack and quick release to capture the vocal from the peaks - then follow it with an upward compressor.
The upward compressor will increase the level of the vocal from the noise floor up - increasing the vocal’s detail. This results in a very balanced and detailed vocal without the need for excessive downward compression.
If you want, saturation could come before upward compression so that the compressor enhances its effect - but it works just as well after. With the saturator, we’re going to establish the desired tone of the vocal - for example, transistor saturation for warmth, or tape from more clarity and mids.
Each saturator is a little different so try multiple until you find one that works well for the vocal.
When introducing temporal effects like reverb and delay, I’m going to do this from the vocal bus. From this bus I’ll create more busses and auxiliary tracks each with one form of temporal processing - so if I want reverb and delay, I’d create 2 busses.
Setting up reverb this way ensures that the delay and reverb don’t interact or affect one another until they reach the output.
If I was to place the delay and reverb on the bus, then the delay would be fed into the reverb or the reverb into the delay - which could be used creatively, but odds are I don’t want this effect.
Once I’ve added all the processing that I want to add to the vocal, and the other tracks and busses in the mix have also been mixed, I’m going to finalize the balance between everything. This can include increasing or decreasing individual tracks or busses.
So say I want the lead vocal to be louder, I’d increase the individual track. But if I wanted the entirety of the vocals to be louder, I’d increase the bus’s fader.