Creativemind wrote: ↑29 Oct 2017
Hi Everybody!
Just found this video on Subtractive EQ on a vocal.
He identifies 3 main issues with the vocal. Muddiness, boxiness and sub-low build up.
What does he mean by boxiness exactly and how did he identify these 3 areas? I didn't really hear anything as such (and I'm listening through 20/20 headphones too.) Does it just take years of practice to just train your ears to be able to identify things to remove?
Also, what things would I be listening for in a vocal overall apart from maybe the 3 he mentions here, anything else? and do people agree with his boosting things at 17mins? he says a little boost to give it presence around 3-4k etc.
What other FX too would people apply to a vocal. Is there a particular chain?
I'm guessing Compress then EQ (as I heard you want to compress before EQ;'ing when mastering so I'm presuming it'd be the same in the mix stage?), then De-Ess and Reverb last?
It's all relative. If you have a vocal that's already harsh, then boosting 3-4k may make it downright painful. I would identify more vocal issues: harshness comes to mind, as does sibilant, overly bright, etc.
Basically, for every frequency range, let's say for every octave, there are good qualities and there are bad qualities. Which implies you can't say any particular frequency is 'good' or 'bad'. Making matters worse, some of these negative attributes are dynamic, which is to say they are not static and you won't hear them at all times. And some may even be contextual - the slightly boomy vocal may work fine if acoustic guitar is all it's up against, but in a dense mix it would never fit.
Some of these can be fixed at the source, which is always desirable when possible. Getting too much proximity effect can lead to some of the low frequency issues, and can be fixed by either backing off the mic or using a different mic. Some microphones pronounce sibilance, and switching mics can help there too.
It can take years to hear these things at first listen, but there are also processes you can go through to help identify them as well. These can include cutting (or boosting) different areas and listening to see how it affects the mix. This can help 'train' your ear so they become easier to identify with experience. It may be helpful to start working through the spectrum by octaves, with a parametric EQ set to cut/boost around an octave wide.
Imagine ten equally spaced bands across the audio spectrum - or better yet just use the Spectrum Analyzer in Reason to do so, as it is (as are most SAs) nicely divided up into ten "octave" bands. For a vocal you can skip the lower few octaves (unless you see something there and want to hear what it sounds like).
Since very vocal is different, it's not a totally straight forward task, but just jump in and start getting your hands dirty - the more time you spend listening and exploring/experimenting, the quicker you'll learn!