When mixing vocals, use some unique techniques like multiband upward compression, or series de-essers to make your vocals sound better. One of the easiest ways to make a vocal sound better is to use multiband saturation, with which you can distort each frequency range for a specific purpose.
For this video, none of the chapters are in any particular order, but they can be combined when mixing vocals. Also, so of these are more on the creative and experimental side, so use your ears and try these out for yourself.
At the moment, there aren’t any multi-band upward compressors other than the free OTT plugin, so what I’m going to do is send my vocal to 3 different busses. On these busses, I’ll place linear phase EQs and then isolate the lows, mids, and highs respectively.
Now I can insert low-level compressors on each frequency range, upward compress heavily, and then blend these signals in with the original vocal.
Let’s listen and notice how this makes the vocal sound really full, and provides an opportunity to equalize the vocal with these 3 bands.
We often use series compressors, but we rarely carry this concept over to de-essers; so let’s try a couple of de-essers out on a vocal. Since they both have unique algorithms, they’ll capture the sibilance in slightly different ways, letting us more accurately capture high frequencies.
Let’s listen to these 2 de-essers on a vocal, and notice how they work together to capture sibilance.
When equalizing, we often use a parametric EQ because it gives us a better idea of what we’re affecting - but we miss the subtle timbre that a vintage equalizer could provide. So let’s start equalizing our vocal with parametric, but then transfer so decisions over to a vintage EQ.
We can use the free plugin Bertom Frequency analyzer, to see the curves of a vintage emulation, and better match up its curve to the parametric EQs.
Let’s listen to a vocal equalized with a parametric eq, then one with a vintage emulation that’s been matched to the parametric settings, and notice the minute differences between the two.
If we’re not careful, this trick can make our vocal sound cacophonous, but with the right settings, following a short reverb with a very clean delay has a really unique and unnatural effect. We’ll use a parallel track, then set up a short reverb, typically under 1 second.
Then, we’ll follow it with a delay with medium feedback, and very distinct delay taps. I find triplet 8th notes or dotted 16th notes work well. Lastly, we’ll blend in the parallel signal. Let’s listen to the effect and notice how it doesn’t sound exactly like either reverb or delay.
Use a multi-band saturator to achieve different timbres and harmonics from different frequencies of your vocal. Depending on the vocal l I’ll create 3 to 4 bands, and pick distinct forms of distortion for each - for example, warm tube is great on the lows and will emphasize transients.
Clean tube on the highs helps brighten up the vocal while warm tape will help to soften highs.
Let’s listen to multi-band saturation and notice how it can be tailored to emphasize very specific aspects of the vocal.
Most delay plugins have some form of an EQ section, but if we set up a send, and use an advanced parametric EQ before delay, we can get creative with what the delay receives. Using the Pro Q 3 I’ll cut out lows while making my mids dynamic.
This way the certain frequencies are emphasized in a program-dependent way.
If we know the key of our song, we can emphasize it with an EQ or a dynamic EQ - this way the fundamental frequency becomes easier to hear and the vocal sounds more in tune. Next, we can add a saturator to create harmonics of the fundamental, which strengthens the effect.
The more we emphasize in-key frequencies, the more in tune the vocal will sound. Let’s listen and see if the vocal subtly sounds more in tune.
So far in the previous chapters, we’ve discussed ways to make a vocal sound better, but if you want to create a lo-fi sound, here’s what you do. Start with an EQ, attenuate above 8kHz and below 200Hz., and boost 2.5kHz - then introduce broken analog equipment emulation.
Since we’re heavily distorting we won’t need compression, nor will we need to de-ess since those frequencies were attenuated by the EQ. Let’s take a listen and notice that only 2 plugins are needed to create a lofi vocal.
If you like the sound of 90s pop vocals, try this at the start of your chain - use an additive EQ and boost 2.5kHz, and 12kHz and up by a few dB. Then heavily de-ess, making the de-esser work a little harder than normal, and achieving a compressed sound.
Let’s listen to it and see if it’s reminiscent of that era of pop vocals.
Using a multi-band dynamic processor, we can both subtly gate and expand our vocals. With the Pro-MB I’ll create my bands, then use expansion with a negative range to gate some low mids, and expansion with a positive range to expand high mids.
This creates a more dynamic sound than simple equalization, while still letting you shape the vocal’s frequency response. Let’s listen to the effects, and pay attention to changes in the frequency response.