peaks... resonance?

[sambrn]

hi,

i usually got well shaped spectrum of my mixes, because i work hard on the equalization of each track at the mixing stage. However while the mid-high (above 1KHz) and the low (under 200Hz) frequencies appear always very smooth, in the 200-600 Hz range i dectect in some cases different peaks (about 5 db).

Monitoring each track with the Voxengo Span free plugin i have discovered that the vocal tracks are responsible for the peaks in the overall mix.
I'm used to compress hard my vocal tracks and to equalize them accordingly to the other instruments in the mix. Trying to attenuate the voice at the frequency where the relative peak appears doesn't affect too much the global result and if i try to cut really hard then the voice sounds bad.

I suppose that the problem is caused by resonance during recording (i'm already working to avoid it in the next rec sessions), doesn't it?

However is there a way to limit these peaks directly on the vocal tracks? Maybe using a multiband compressor (like Waves lin eq)? Suggestions?

Excuse me for my english.

Sam

[Ged the undead]

It's not a BAD thing that there are peaks in the midrange due to your vox tracks.

I'ts perfectly alright, i mean it may LOOK bad, but it's all about, does it "sound" bad?

Music is supposed to have loud parts and quiet parts remember,
don't get caught up in the "visual thing"
listen to your track when the vox comes in,
Does it "sound" wrong? if so why? EQ issiues with the actaul vox? or pehaps better compression settings on the vox would glue it in the mix better.

There will always be peaks here and there on the spectrum, if there were not, it would sound rather unatural and boring.

Only during the mastering should you even bother about smoothing out the balance, even then if you liked your mix to start with and it plays well on most systems you should not need to do much EQ work.

I always go with the "average" spectrum in mastering, if there are peaks on the peak spectrum, then thats what i use freq dependant compression for.

Hope that helps mate.

[sambrn]

Thanx.

No, the mixes usually sound good even when there are these peaks in the mid range. As i said, i'm used to make good eq on each track during mixing process than rather make great adjustements on the mix.

The reasons that brought me to the question are:

1) Scanning the spectrum of commercial songs I don't often find peaks like the ones i got... the spectrum is smooth everywhere (for example i've scanned some tracks from Diamonds on the inside of Ben Harper)

2) why in the mid-hi region can i achieve a smooth spectrum and in the low-mid no?

However you answer sounds very good to my ears. It helped me.

Sam

[Ged the undead]

Thanx.

No, the mixes usually sound good even when there are these peaks in the mid range. As i said, i'm used to make good eq on each track during mixing process than rather make great adjustements on the mix.

The reasons that brought me to the question are:

1) Scanning the spectrum of commercial songs I don't often find peaks like the ones i got... the spectrum is smooth everywhere (for example i've scanned some tracks from Diamonds on the inside of Ben Harper)

2) why in the mid-hi region can i achieve a smooth spectrum and in the low-mid no?

However you answer sounds very good to my ears. It helped me.

Sam

The spectrum basically represents nothing but level and frequencies.
Remember, your comparing your MIX to an already mastered track.

What you should try and do is smooth the balance a little in harbal till it sounds where you want it, then when applying compression, either boost the sidechain eq of the said vox frequencies or use a multiband comp on the vox area.

Then, import the mastered track into harbal and see if that region is a little smoother.
Of course, if it sounds smooth on your system and others you try it on then it's fine, EVEN if you still see a peak in that area.

I've seen countless refrences that have peaks etc but some folk forget that the spectrum only tells you so much, and forget that most likely some freq specific compression was used.

cheers,
Ged.

[sambrn]

Thanx,

i will try soon to do what you suggested me. I have always tried to apply a multiband comp to the vocal track (in the frequencies area of low-mid) and i have obtained a 5 dB reduction of the relative peaks without having particular changement in the global vox sound and neither in the rms spectrum. I will try soon to mix with this comp on vocals channels and then check the result with har-bal.

Regards,

SaM