To make vocals sound better, start by editing - when editing, cut out background noise, use clip gain to balance dynamics, and even remove plosives or unwanted sibilance. Then you can introduce subtractive EQ, Compression, Additive EQ, Saturation, Exciters, and various forms of short and long reverb.
Before we begin to process our vocal let’s clean up certain aspects of the performance with some editing - we’ll reduce cut out any aspect where the vocal isn’t present to reduce background noise. We’ll then use some fade in and outs to clean up where we made these cuts.
Next, I’ll use a little clip gain to make the performance dynamically balanced.
If you have the program, I’d also recommend using declick on your vocals with RX - this gets rid of unwanted mouth sounds that we don’t want to amplify later on.
I’m going to use an EQ to remove 3 sections of the vocal, the first is the nasal tone which is caused by vibrations in the nose and nasal cavity getting picked up by the mic we’ll dip about 700Hz to get rid of this.
We’ll then use a high pass filter to attenuate plosives, and then find the sibilance in the high range and dip that with a higher Q bell filter.
If you have a dynamic EQ, make this cut to the sibilance dynamic so that the cut only occurs when messes are present.
Next, I’m going to use a compressor to balance out my vocal’s dynamics - I’ll use a quick attack and moderate release of about 30ms to capture the vocal quickly then release it quickly so that the details are retained. I’ll start with a 2:1 ratio and lower the threshold.
I’m trying to achieve between 3 and 6dB of attenuation so that the compression is controlling dynamics, but isn’t too noticeably changing the vocal’s timbre.
With an EQ let’s focus on 2 areas of the vocal - first, let’s find the voice’s fundamental around 200Hz and subtly amplify this. If the vocalist has a lower voice, this fundamental may be lower, and the opposite is true if the vocalist sings or speaks in a higher pitch.
I’m going to boost this by about 2dB which will add some warmth to the vocal. Now, I’m going to amplify a higher-ordered harmonic of the vocal around 2.5 to 3.5kHz. Listen intently to find the note in the range that’s in key with the song, and boost that.
Mild distortion or saturation will add harmonics to the vocal, which reinforce the fundamental frequency as well as fill in some of the gaps in the mid frequencies. If you want your vocal to sound bright, try an exciter that adds harmonics higher up the spectrum.
For more warmth use a transformer saturator to get some good low-end amplification. Tape saturation is usually great for the mids since it creates a strong 3rd order harmonic.
Following our last thought about exciters and high-ordered harmonics, we can combine an exciter with an airband EQ to create an incredibly bright or clear-sounding vocal. This is better for more musical applications than for dialogue since it creates a somewhat unrealistic sound.
First, insert an exciter to generate high-ordered harmonics - then, insert an EQ with an airband and amplify the high frequencies. If your vocal has excessive sibilance, be sure to use a de-esser before your exciter.
For a more powerful sounding vocal, we need to create short reflections which will cause a dense and impressive sound - since the reflections or delays are so short, our mind perceives the signal as coming from one source. Instead of a delayed or reverberated vocal, we perceive one powerful vocal.
Keep short reverbs under 500ms and short delays under 120ms to accomplish this effect.
After you’ve made your vocal sound thick with short delays and reverbs, let’s use some longer utilized ones to blend the signal in with the surrounding instrumentation. I’m looking to create a medium-length reverb, that focuses on the mid-range, and is in time with the song.
I’ll use a preset from the FabFilter Reverb plugin, that emphasizes reverb on the mids, and then I’ll divide 60000 by the song’s BPM to get 1 quarter note. I’ll set the RT60 to a multiple of this time somewhere between 4 or 5 seconds.
With the wet/dry I can blend this effect in, or I can introduce this effect as a send.
There are some occasions in which limiting can be useful when mixing vocals, like when processing vocals for a podcast. When using limiting on a vocal, it makes the most sense to use as clean of a limiter as possible, as a means of reducing distortion.
For example, the modern algorithm on the Pro-L2 doesn’t generate harmonics unless under extreme settings. But if I was using an algorithm of limiter designed for pop or rap, I’ll get these harmonics.
One trick you can use if your vocal’s transients are sounding a little too harsh, turn on true-peak limiting, which will gently reduce their transients.