Professional Analog Mastering.
Professional Analog Mastering.
226K+ subscribers
Before we add processors, let’s determine how loud we want the vocal to be, and how loud we want the beat to be.
Ideally, they’ll be equal in amplitude, although you can vary that slightly as you see fit.
On the unprocessed beat and unprocessed vocal, I’ll insert an LUFS meter.
This free meter from GoodHertz is a great option.
I’ll let the track play while the meters measure the integrated loudnesses of each. Once they’re measured, I’ll either amplify or attenuate with clip gain to ensure they have the same loudnesses. This GoodHertz meter has the option to match the level to a particular LUFS, but I want to remove this plugin, so I’ll just keep the figure in mind, and then make the needed changes to the clip itself.
It’s a simple but often overlooked step - knowing how you want the vocal and beat to interact loudness wise will help you make more informed decisions as you introduce various processors.
Let’s take a quick listen to the vocal and beat having their loudnesses matched.
Watch the video to learn more >
This step is arguably boring, but it needs to be done for the best sound possible.
When editing, I prefer an FFT editor like this RX platform by Isotope. I don’t recommend plugins or software that often, but an editor like this is incredibly valuable to have - even if it’s the introductory version.
Its gain feature allows for incredibly accurate equalization, which I’ll use to create a reasonable level for the sibilance, attenuate any plosives or pops, remove any unwanted noise or electrical pops that occurred during recording, and so on.
To use it, highlight the range you want to affect - in this instance I’m isolating some aggressive sibilance that I want less of. Once it’s highlighted, I’ll selected the gain module and reduce the amplitude of the region by a few dB, or as much as needed.Don’t overdo this, otherwise the performer will sound like they have a lisp.
Then, I’ll find the vocal’s fundamental frequency - this is indicated by high amplitude notes, usually between 100-200Hz.
Once I’ve found that, I’ll highlight the range that’s below it - then, select the gain module, and completely remove these frequencies.
Everything below the fundamental is either noise, rumble, plosives, or any unwanted and unmusical aspects of the performance - just make sure you’re not affecting the fundamental in any way.
Once I’m done I’ll overwrite the original file, which will replace it automatically in the DAW.
Alternatively, you could use clip gain to control sibilance by its a lot less accurate. You could also use EQ and de-essing to control unmusical aspects, but again, they’re less accurate.
It’s probably not popular telling you to use a somewhat obscure platform to perform editing, but, I’m being completely honest here, it makes a huge difference in whether the vocal sounds professionally controlled, or like it didn’t receive the attention it needed.
Let’s listen to the vocal before and after this editing.
Watch the video to learn more >
So, I’ve inserted an EQ on the Vocal and an EQ on the beat.
What I want to do is shape the 2 around one another so that they blend naturally.
This will look a little different from song to song; however, dipping a little of 250Hz on both the beat and the vocal will help introduce some clarity.
Meanwhile, a small boost around 3kHz on the vocal will help it sit forward - something we may need to account for later if it alters the loudness too much.
If the vocal as some nasally tones I could decrease somewhere between 700-1.3kHz, depending on where the resonance is. Meanwhile, I could slightly boost this on the beat to help it mask some of that tone.
While we’re here, we might as well control some of the stereo image of the beat.
A mid-side EQ is great for this - I’m using the Pro Q 3, but ZL equalizer is a fantastic free alternative.
With it, I’ll attenuate some of the beat’s side image up to about 80Hz with a HP filter - again, with its placement set to the side image.
Then, I could boost a little of 400Hz on the side image to make it sound fuller without creating a muddy or boomy sound.
Lastly, so high frequency expansion sounds great - a high shelf on the side image does this well.
If the kick and bass are lacking, I could use a mid image filter to amplify their fundamentals.
If the snare is getting buried, I could find its fundamental or an overtone and boost that slightly.
Just keep in mind that any change you make will greatly impact the sound, since we’re dealing with the full stereo beat here.
Let’s take a listen to these changes. I’ll compensate for any loudness variations to keep the 1:1 relationship between the beat and vocal’s loudness.
Watch the video to learn more >
So, I’ve covered this chain before, but in greater detail in a video titled how to create crisp vocals - I’ll link it below if you want an in-depth explanation. But I’ll quickly cover it here.
After the EQ, if needed, use a de-esser to control the sibilance, although the editing we did should have that covered.
Then, introduce a compressor with a 4:1 ratio, a softer knee, a quick attack, 50ms release, and a couple milliseconds of lookahead. Try to achieve about 6-8dB of attenuation. I’m using the Pro-C2 but a great affordable alternative is Toneboosters Compressor V4.
Automatic make up gain combined with these settings will bring a lot of the vocal’s quieter details to the front.
If needed, use a vocal tuner at this point - after the compressor to help it achieve a more accurate reading.
Then, introduce some mild saturation, ideally tube saturation, to give the vocal a slightly more aggressive and biting sound.
Next, I’m going to create 2 sends from the vocal; one will compress the highs, and the other is just a ducked reverb send.
Starting with the compressed highs send, first insert a linear phase EQ and isolate the highest frequency range. Insert a compressor next and compress heavily. Lastly, introduce a subtle reverb - one with airy highs will work well, and use a low mix amount for the wet/dry.
Moving on to the 2nd send, we’ll introduce our primary reverb. Use whichever reverb you like and you think works well with the track, and set it’s wet/dry to 100%. Then, insert a compressor and set the original vocal track as the external side-chain.
Set quick settings, but avoid any automatic make up gain.
This will compress the initial part of the reverb, reducing its level during the vocal’s transient. This helps to retain intelligibility.
Alternatively, you could just use pre-delay, but this just pushes the start of the reverb back. The ducking we’ve set up here causes less reverb on the vocal’s transient depending on how aggressively we’re compressing. It sounds a bit more natural in my opinion and gives the vocal a cohesive sound.
Lastly, blend these 2 aux tracks in with the original signals via their faders.
If you want to control the original vocal and these 2 sends collectively, change their outputs to the same bus, and all 3 will be routed to that new bus.
Let’s take a listen to the processed vocal, and notice how it’s a lot closer to a polished sound than before.
Watch the video to learn more >
When you’ve found a beat or someone made it for you, it’s more or less already mixed. Odds are it’s gone through some limiting and maximization as well.
Regardless, it doesn’t hurt to subtly maximize it a little more, and maybe introduce some limiting as well.
The processor you pick is up to you, I’m going to use the enhance function of this Sonnox Limiter, and then the Pro L 2 to push it higher, while allowing some of the true peaks to get across. This should help it sound a little louder after normalization.
If you’re curious as to why, check out our video about making a track loud after normalization.
Then, I’ll insert a maximizer on the vocal bus - again, this includes the original vocal and the 2 auxiliary or parallel tracks we created.
I won’t limit it, just increase the quieter details to bring up the overall loudness. At this point, if you need to make adjustments to the compressed high frequency range aux track, or the parallel reverb track, you can definitely do that.
If you want to include a final EQ on the stereo output to shape everything at once, you can do that, but you’ll likely need another limiter to protect from overs.
Instead, I’ll just make the needed EQ changes to the vocal and beat separately to achieve the final sound I want.
Lastly, I’ll check one more time to ensure the 2 signals have equal loudnesses.
Let’s take a listen to the finished mix and quick master - how you approach this process may be a little different, but this is a great way to create fantastic results.