To begin this section, we need to say that problem frequencies can move around as a vocal moves from note to note. Therefore, you will also affect multiple notes differently with the cuts you make, as the EQ bands stay in place while the vocal moves around.
You need to trust yourself and decide if you want to automate the bands to move along with the problematic frequencies, use a dynamic EQ, or keep a static cut in the EQ. However, for simplicity's sake, we will stick to static EQ cuts in this section.
Furthermore, with deep lows in a vocal, it is recommended to cut them out. Even if the vocalist does not reach this range, the mic may pick up external noise, eventually overlapping elements like the bass or synths in the final mix.
Furthermore, make sure not to use a broad Q or bandwidth with a high pass filter, as over-cutting can sterilize a mix, making it sound less professional. Instead, a slope of 12 to 18 dB/octave should be acceptable.
Additionally, most other cuts should be done only if a specific problem needs to be solved. For example, if you can hear a problem but do not know where it is in the frequency spectrum, create a bell filter with a high Q value and boost it quite high. Then, sweep this boost from side to side slowly.
You are welcome to boost or attenuate as much as you want, but be sure not to go over the top. Also, keep using professional tracks for reference or as a North Star to guide your mixing and mastering decisions.
You might often hear a vocal described as nasally, which generally means there are resonant peaks between 500 Hz and 3 kHz. This quality is often just an attribute of the singer's voice, but it can typically be handled quite easily.
Since we have previously discussed ensuring a solid foundation from the get-go, the correct mic placement can help tackle this issue without needing EQ.
So, try to get your vocalist to sing off-axis and A/B the results.
Another issue that can arise when EQing vocals is sharp sibilance. Sharp sibilance is a term used to categorize consonants that produce a hissing sound, such as s, sh, ch, z, zh, and sometimes even t.
Unfortunately, a pop filter will not tame these, as they are often meant to plosives; sounds like b, d, g, p, and sometimes even t and k. However, you can use the bell filter technique to find the sibilance and attenuate it, but we will discuss de-essing later.
Moreover, a compressor will often follow the subtractive EQ in a vocal processing chain. However, the compressor will probably note some of the cuts done in the EQ, so it is up to you whether you keep tweaking these settings to get the desired balance or compression beforehand.
Nevertheless, neither approach is wrong. Depending on your vocalist and track, you will have to experiment to see which yields the best results.
Professional Online Mixing and Mastering Services