Although just about all compression will control your dynamics, specific advanced settings will greatly control the level of your vocals. Start by using a higher ratio of 5:1 and combining it with the shortest attack time possible - this will ensure that the totality of the vocal is being compressed.
Then increase your input by a few dB, and reduce the output by a similar amount, driving the vocal into the compressor while compensating for the gain.
Then, use a medium-length release, around 300 milliseconds - this holds onto the vocal for long enough to control it, without losing vocal intelligibility.
If you really want to squeeze your vocal, combine optical compression with a soft-knee setting and a low threshold. This will cause gradual but significant compression, which won’t be super audible while still keeping the vocals within a more-or-less 10dB dynamic range.
First, choose an optical setting for the compressor - then create a soft-knee and lower the threshold until you’re achieving up to 8dB of compression.
Your ratio should be between 2:1 and 4:1, depending on how much control you need. Notice how the compression occurs very gradually, and in a nonlinear way.
Quieter parts get compressed less, and louder compressed more.
Compression can be used to de-ess your vocal in a more complex way than your typically de-esser. Whereas a de-esser offers only a few controls, when you de-ess with a compressor, you get just as much control as you would with your regular compressor.
To do this, you’ll need to use a compressor with an internal side-chain option - like the FabFiler Pro-C 2 or the Weiss DS1 MK3.
Using the internal side-chain, isolate just the sibilance-based frequencies, which are usually between 5 and 10kHz, but have the potential to be a little lower or higher.
Then introduce compression as needed.
Some compressors make compressing vocals easy - they offer algorithms designed specifically for controlling and augmenting lead vocals and BGVs. Two compressors that I’ve found which do this well are the FabFilter Pro-C2 and Softube’s Drawmer S73 - which offer both vocal algorithms and some great presets.
The FabFilter compressor utilizes an automatic knee setting and an automatic ratio, both of which become program-dependent and alter depending on the incoming vocal.
To select it, simply find it in the Style section of the compressor.
With the Drawmer S73, the vocal option is designed to bring a vocal in a mix forward, but it can also be used on an individual vocal.
Although it may sound counterintuitive at first, using a limiter on a vocal can be really helpful at controlling dynamics without introducing distortion. If you find that you need aggressive compression to control a vocal, but still want it to sound clean, try using a limiter.
The reason why limiters work so well in this situation is that they’re designed to handle the full 20 - 20Khz frequency range aggressively, but with minimum distortion.
For example, the Modern setting on the FabFilter L2 is a great option, or you can simultaneously reduce the threshold and output of the Waves L1 and get a similar effect, and possibly even better effect.
If you want a particular frequency or set of frequencies to be compressed slightly hard than others, you can use an EQ before your compressor to emphasize them. If you emphasize frequencies before compression, it can create a very unique tone that you won’t have been able to achieve otherwise.
For example, I may want to boost 2kHz to add clarity to the vocal, and then have those frequencies compressed harder. This will balance out the boost, while simultaneously causing more compression coloration and tone.
The effect is somewhat subtle but can be used to great effect when compressing vocals.
If you don’t want a compressor to affect a particular frequency or to add its timbre to that particular part of the vocal spectrum, you can de-emphasize with an EQ prior to the compression occurring. Additionally, you can add these frequencies back in with another EQ after compression.
For example, say I don’t want a significant amount of compression to occur to the 2kHz range - or in other words, I don’t want the timbre of the compression to greatly impact the 2kHz range.
I’d use an EQ to emphasize that range, then compress the vocal. If reducing those frequencies made it sound unbalanced, I could use a second EQ after the compressor to add them back in.