If you’re trying to use less overall compression, probably because you want a cleaner sounding vocal, use clip gain on your vocal first. Although time-consuming, find all aspects that are too quiet or too loud, isolate them, and then alter their respective gains accordingly.
Keep in mind that some frequencies, even when at a lower level, will sound as loud or louder than ones at a higher level, so you’ll need to use your ear.
Let’s listen to a before and after of clip gain being applied.
Certain compressor settings are great for getting that slight after upfront and highly detailed sound - to accomplish it you’ll need to capture all of the vocals with your compressor. Use the quickest possible attack, a 50ms release, while enabling about 2ms of lookahead, and auto-make-up gain.
This way the compressor can read and be triggered by the full signal, then amplify the compressed signal to match its original level, bringing quieter parts of the vocal to the forefront. Let’s listen and notice how close the vocal sounds.
Although the title of this chapter says 2kHz, it’s more like find the vocal’s harmonic that’s around 2kHz, and boost that with a bell filter by 3 to 5dB. Then, insert a compressor and compress the vocal as you normally would, or maybe how we did in the previous chapter.
By driving this frequency into the compressor, it works a little harder on it, causing a unique timbre and emphasizing what makes the vocal cut through the mix. Let’s listen and notice how the vocal’s presence increases.
In short, yes - saturation causes soft-knee downward compression at higher input levels and reduces the dynamic range by introducing harmonics. Different saturation types have different compression knees, ratios, and harmonic formations, making saturation a truly complex effect with a lot of variables to consider.
For example, tape saturation compresses significantly more than tube saturation, whereas transistor saturation mildly compresses, but then compresses heavily at higher levels.
Let’s listen to saturation at higher levels on a vocal and see if we notice compression.
Although there’s no reason that you can’t use both, it helps to know what’s different about how saturation compresses and how a typical compressor compresses. The primary difference is that saturation will compress different frequencies to varying degrees, whereas a compressor will compress the same for all frequencies.
For example, a compressor will attenuate 200Hz and say 10kHz the same amount, unless it’s an optical compressor, but a saturator may compress the lows more, and highs less, or vice versa, depending on the saturator type.
Let’s isolate the lows and highs and saturate them with the same settings, to see if we notice a subtle difference in the amount of compression.
I’ll admit that I just made up this term, but it’s worth noting that a typical saturator introduces harmonics way before it begins to introduce downward compression. If we reversed this process, we could use a compressor to attenuate the signal, and then introduce the harmonics with a distortion plugin.
As a result, harmonics form off of the dynamically controlled signal, and subsequently aren’t subjected to further compression, resulting in a compressed vocal, but with a unique timbre. Let’s listen to it.
Earlier we discussed how saturation compresses different frequencies to varying degrees, which arguably gives it a more complex sound - but what if we could achieve this while being able to exercise a lot more control over the signal? To do this I’ll first need to duplicate my vocal.
Then I’ll place an EQ on each vocal track with one isolating the lows and the other isolating the highs, both set to low-latency linear phase to retain their timing.
Then I’ll compress the lows one way and the highs another - letting me determine how much each range is compressed, the overall timbre of the compression, and more.
Granted you could do this with a multi-band compressor, but many lack the complex settings a standalone compressor would have.
Let’s listen to this setup can consider how it could be used both practically and creatively.
We discussed downward compression a lot, but upward compression is just as useful when mixing and compressing vocals. I’ve covered this before, but an upward compressor will capture quieter details, maybe about 20 to 30dB below the peak, and amplify them while leaving the signal’s peaks unaffected.
If we combine upward with downward compression, we control the dynamics of the vocal from both directions. Let’s listen and notice how the vocal sounds full and impressive.
Another good compression pair for vocals is peak down compression and then peak down optical compression. The first compressor will have quick settings, similar to those described in chapter 2, then an optical compressor will smooth the remaining dynamics with its dual-timed and frequency-specific release.
Combined they create an upfront and controlled vocal, but with a smooth and enjoyable sound. Let’s listen to it.
Although this seems incorrect, short reverb can be seen as a form of upward compression, since we’ll perceive these short reflections as coming from one sound source. With that in mind, let’s use a very short reverb, lower the pre-delay, and isolate the reflections to the mids.
With it blended in, we’ll notice that the vocal sounds thicker and fuller - this is due to the added low-level content and the reduced dynamic range caused by the newly added reflections. Let’s listen to it.
A vocal’s temporal effects typically include reverb, delay, and maybe chorusing and modulation effects - as we covered, short reflections reduce the dynamic range by adding in low-level content that’s perceived as coming from the original source. If we compress these reflections, we enhance this effect.
To show this, let’s set up a send on which we’ll use a short reverb, followed by a compressor with auto-make-up gain. Then we’ll blend in the effect and notice how these compressed reflections really make the vocal sound thick.