19-05-2015, 11:15 PM
These are just my opinion, lots of people swear by dynamic EQs and multiband compressors for this, but I disagree mainly because of how long it takes to get it right without really damaging the vocal sound.
Sometimes something as simple as firm full band compression on a vocal track can help even out resonances, allowing you to use gentler static EQ, post-compression, to attenuate the low registers without unevenly compressing the track. I've personally found dynamic EQs or multiband compressors to be unacceptably time consuming to be transparent for this purpose unless the resonance is quite severe, in which case dynamic EQ is often the only thing that will contain it. For me the lower register of the lead instrument is too important to process severely unless you're going for an airy pop vocal, so I'd personally avoid that approach if possible.
It's also quite often in the nature of a particular singer's voice to resonate in various places, so it might not be a problem at all! You may well be blunting what could be a unique characteristic of the voice, and that'd be a bad thing in this style (again, my opinion.)
Another thing you can try is to decide which instrument in the mix is primarily living in the general region that the vocal is resonating, quite often the bass guitar or snare drum. Those instruments will also share the center of the stereo image, so in those cases the resonance will be problematic. A good strategy there is to route both through a buss and apply gentle, low ratio compression... not so much that it will affect the dynamics of either in the mix as a whole, but enough that the compressor causes vocal and instrument to react to each other dynamically. Again, unless the resonance in question is very severe, the right compressor at the right settings will in effect "glue" the two tracks together, therefore gentle EQ on the individual channels at the offending frequency will then take on a time-domain qualities relative to the buss compressor... so you get a kind of really gentle dynamic EQ on -both tracks- simultaneously, relative to one another. In the days before dynamic EQ and accurate multiband compression existed, that's probably how those time-related disagreements were resolved, since mixing on an analog console requires bussing by its very nature.
If you can get characterful aspects of multiple instruments to subtly shape one another, it's good all the way around in my opinion, because it naturally results in a more dynamic sound and will reduce the need for blending reverb and super-meticulous automation in final stages of the mix. This strategy is becoming a very important part of the way I mix, especially when it comes to getting the drums and bass guitar to work with each other constructively.
Anyway, my main point is that unless you're working on a song with a really dense arrangement or a genre where transparent vocal processing isn't necessarily best... I'm talking about chart pop, dance, electronica, urban... fiddly processing on the vocal is going to be too time consuming (and often creativity destroying) to be worth it, especially when there are much easier, gentler methods. Frequency dependent compression is really best reserved as a last resort on primary mix elements, not because it's ineffective, but because a simple combination of full band compression, EQ and creative routing on the way to the master buss will get similar results and take a lot less time... and that means you'll have more time for creative, artful mixing.
Why mangle your tracks when you can allow their inherent attributes to cooperate and positively contribute to the mix? You'll certainly stand out from the crowd. Just my two cents.
Sometimes something as simple as firm full band compression on a vocal track can help even out resonances, allowing you to use gentler static EQ, post-compression, to attenuate the low registers without unevenly compressing the track. I've personally found dynamic EQs or multiband compressors to be unacceptably time consuming to be transparent for this purpose unless the resonance is quite severe, in which case dynamic EQ is often the only thing that will contain it. For me the lower register of the lead instrument is too important to process severely unless you're going for an airy pop vocal, so I'd personally avoid that approach if possible.
It's also quite often in the nature of a particular singer's voice to resonate in various places, so it might not be a problem at all! You may well be blunting what could be a unique characteristic of the voice, and that'd be a bad thing in this style (again, my opinion.)
Another thing you can try is to decide which instrument in the mix is primarily living in the general region that the vocal is resonating, quite often the bass guitar or snare drum. Those instruments will also share the center of the stereo image, so in those cases the resonance will be problematic. A good strategy there is to route both through a buss and apply gentle, low ratio compression... not so much that it will affect the dynamics of either in the mix as a whole, but enough that the compressor causes vocal and instrument to react to each other dynamically. Again, unless the resonance in question is very severe, the right compressor at the right settings will in effect "glue" the two tracks together, therefore gentle EQ on the individual channels at the offending frequency will then take on a time-domain qualities relative to the buss compressor... so you get a kind of really gentle dynamic EQ on -both tracks- simultaneously, relative to one another. In the days before dynamic EQ and accurate multiband compression existed, that's probably how those time-related disagreements were resolved, since mixing on an analog console requires bussing by its very nature.
If you can get characterful aspects of multiple instruments to subtly shape one another, it's good all the way around in my opinion, because it naturally results in a more dynamic sound and will reduce the need for blending reverb and super-meticulous automation in final stages of the mix. This strategy is becoming a very important part of the way I mix, especially when it comes to getting the drums and bass guitar to work with each other constructively.
Anyway, my main point is that unless you're working on a song with a really dense arrangement or a genre where transparent vocal processing isn't necessarily best... I'm talking about chart pop, dance, electronica, urban... fiddly processing on the vocal is going to be too time consuming (and often creativity destroying) to be worth it, especially when there are much easier, gentler methods. Frequency dependent compression is really best reserved as a last resort on primary mix elements, not because it's ineffective, but because a simple combination of full band compression, EQ and creative routing on the way to the master buss will get similar results and take a lot less time... and that means you'll have more time for creative, artful mixing.
Why mangle your tracks when you can allow their inherent attributes to cooperate and positively contribute to the mix? You'll certainly stand out from the crowd. Just my two cents.
I'm grateful for comments and suggestions. Thank you for listening!