When trying to make your vocal clean, it can help to use Izotope's RX to remove vocal clicks and potential clips. Additionally, you may want to use clip gain to reduce plosives, use notch filters to remove hum, and even convert your vocal into MIDI.
One of the hardest things to remove from a vocal recording is vocal clicks, or the sounds made by a singer’s mouth - to get rid of these I’d recommend opening Izotope RX and using their De-click module. Make the detection random, emphasize measurement in the high frequencies, with low sensitivity.
You can then preview the processing and monitor which clicks are getting removed from the vocal.
If you received a vocal recording in which clipping is a problem, say the engineer hit the preamp too hard resulting in unwanted distortion, you can use another RX module to repair some of these clips. Set the right threshold, use the highest quality rendering, and preview the processing first.
Once your signal has been repaired, import it back into your DAW and continue mixing.
If you don’t want to compress your vocal, you can use a vocal rider - it works somewhat similarly to automation in that the full signal is turned up or down depending on the level. You can also sidechain the full instrumental mix to have the vocal respond to its level.
This way you avoid some of the artifacts and timbre altering aspects compression can cause.
Plosives can typically be attenuated by using a high-pass filter, but if their frequency goes higher than you’d like to cut, this can be an issue. In this case, you’ll need to find the plosives in your vocal track, isolate them, and attenuate them using clip gain.
This ensures a plosive free track that still retains its low-frequency response.
Hum is usually located around 60Hz, but unfortunately, it often includes harmonics that exist higher up the frequency spectrum. If your recording has a hum, you’ll need to create multiple notch filters to remove the hum - in this example, 60Hz, 120Hz, 180Hz, 240Hz, and possibly higher.
When you have to use notch filters on your low frequency, be sure to use a linear phase setting to avoid aggressive phase shifts to the signal.
In-time reverb and delay are helpful ways to create a clean vocal, but in-time compression can be just as beneficial - use 60000/BPM to get a quarter note in milliseconds. Then divide this number by 8 to get a 32nd note resulting in quick, in-time compression.
It’s best not to use a quicker note than this since you’ll likely distort the vocal’s low frequencies.
Sometimes, you’ll want to aggressively attenuate your vocal’s dynamics and make it as controlled as possible - when this is the case, it’s best to use a limiter with a lowered ceiling than a compressor. Additionally, if possible, time the release of the limiter to the BPM of your track.
Just like with the compression, avoid super quick notes, since this can cause distortion to the vocal’s low frequencies.
This is a somewhat unique and unorthodox trick, but if you convert your vocal performance into midi, you can use that midi as a support for your vocal. With the right instrument, your vocal will sound brighter, a little cleaner, and definitely a lot stronger pitch-wise.
To do this, follow these steps:
If your vocal is sounding weak, but boosting with heavy compression and makeup gain is introducing too much noise or bringing out distortion, try using a chorus effect. Set up a send and insert an EQ with a high pass filter around 300Hz - then insert a vocal chorus effect.
This way you can increase the power of the vocal without resorting to noisy processing.