Follow Up...Vocal Mixing Approach

Question

I've been kicking around possibilities in a different thread with some very helpful people
http://discussions.apple.com/thread.jspa?threadID=2386425&tstart=0
Now, I'd like to implement what AT THIS POINT, I think may be my best way to approach the issue of getting my vocal mixes good.
First, I am going to go back to using my mic the way I had in previous recordings. That is, using the large windshield that came with my SM7, that helped with a lot of plosives etc. Also, I will apply the bass roll off and NOT the presence boost. Presence boost seemed to give too harsh highs. I can EQ later if need be. Flat will be easier to start with. As far as pre tracking, this is about all i can do, since I know my recording levels and all are good, and i don't have any other equipment besides the SM7 and the DUET.
Next, I would like to know y'alls take on "NORMALIZE". I usually have not ever normalized my vocal tracks. The last one I had issues with a lot, I actually DID normalize, So, perhaps I WON'T in the future. I'm gonna open up a meter for monitoring. Prolly the Multimeter. I'm going to go in and for the sake of time, only do some manual edits to the volume envelope for the REAL bad guys, those that stand out clearly in my ears. Maybe do a quick check for anything that is REAL quiet too, boost that a little manually. Problem is I don't know how much to boost or lower. Can I see with the meter what the overall average db is, and see if I have a loud part that hits, say -6db and the average is around -12db, so I'd just lower that part by 6db?? and vice versa for quieter parts.
Now, the fader is controlled by the volume envelope so if I do automation I will lose control of the fader for my levels, so should I insert GAIN plug or just use an output gain on one of the inserts I'm gonna use anyways. Either way, at this point I will set the relative levels. Just a quick setting to start off with.
Now I believe I can apply this plugin
http://sounds.wa.com/audiounits.html
Thanks Ericksimon!!
Looks like a nice tool, similar to what I was looking for but not too cumbersome. I have not gotten a chance to try it out or see how to use it properly.
So at this point, I think I should have a really steady volume level going on.
Now I am not sure about the whole compression thing. I have come across several options. First though, do I want to do any EQing at this point??? before I add compression or after?? There are two points I would want to use EQ, one is to tame any out of control frequencies and 2 would be to shape the overall tone of the vocal. I suppose I could EQ for any bad frequencies first, then do compression and whatnot, then EQ at the end to slightly shape the sound if need be.
Anyways, compression. The Multipressor worries me a bit. I've used it with nice results on instruments and sampled audio, but not on vocals. Also, the downward expansion looks cool. I don't usually have any problem with too much noise, but I guess that's a problem i just never knew I had!!! I just try to track with as little gain as possible while still achieving a usable recording level. Then I get increase the gain later without having a lot of noise on the recorded track. Also, the multipressor would address my Highs in my voice jumping out at times. If I could just compress the Highs and not so much the lower frequencies. That might give me better results than JUST compression the whole spectrum equally.
The OTHER option, maybe, would be to use the DeEsser. I can compress a specific frequency range, but I don't think it has multiple bands. So, I couldn't really get any compression going on for the other bands. But I was thinking about the DeEsser in tandom with a Noise Gate instead of the Mulitpressor. This may be a little easier, but I think I might want to have a full spectrum compression, but be able to compress certain "trouble" frequencies more or less.
So I think I should be done at this point as far as levels and all that go, perhaps if I was to use EQ to color my track a bit, but that is gonna fall in with applying REverb and Delay or whatever else I want to add flavor. WHICH IS A WHOLE nother thread. After mixing about 13 tracks in logic now, I still can't find a setting that I'm comfortable with for my voice. Every recording seems different so every time i have to do something quite different. And every time i feel like it's the first time!!!
Anyways, how does my approach at this point look??
Holy crap,.....I didn't realize I had written that much. My god, If anyone replies you are truly a saint.

Answer 1

Bee Jay wrote:
Next, I would like to know y'alls take on "NORMALIZE".
Never do it. Absolutely pointless.
I think if it doesn't hurt anything I WILL because I think it helps me when doing manual edits so that I have a point of reference that is somewhat consistent throughout multiple audio regions
Now, the fader is controlled by the volume envelope so if I do automation I will lose control of the fader for my levels,
This is a good point - in general, send the output of a channel to a bus/aux channel. You can now automate one, and use the fader on the other to offset the levels - doesn't really matter which, although I prefer to automate the channel fader to smooth out the signal before hitting the aux wher the compressor is.
I am not real familiar with aux channels, I've tried using them but didn't understand the send amounts and how it was affecting the signal. If I put the send amount at 100 does 100% of the signal get passed to the fader then. So in your example, I would definitiely want to set at 100. As of now I simply used the gain control on the compressor or whatever plugin was last in the chain, I think that should not be any different in the sound....
The Multipressor worries me a bit. I've used it with nice results on instruments and sampled audio, but not on vocals
I wouldn't use it unless there is a definite reason to use it. At this point, you don't sound sure why you are using, and thus are unlikely to use it effectively. So don't over-complicate and use something simpler - there are plenty of good sounding compressors around, if you don't like Logic's.
well what I did is looked at the problem frequencies that I had corrective EQ'd and made that one band,,,,from say 700-1200, then made the other two bands as 0-700 and 1200-18000 or whatever. I didn't touch the low band, but I applied compression slightly different for each one of the other bands. I think it turned out much better than I expected, but I really don't know that I needed much compression anyways......the manual automation, rider plug and EQing had the signal going to the compressor pretty good already. It's amazing how much better a compressor sounds when the signal coming in isn't pure garbage!!!
I just try to track with as little gain as possible while still achieving a usable recording level.
What levels do you record at?
I believe the settings on my Duet are at about 56-58 usually when I record. This puts the input signal usually bouncing around between -24 to -12 db. I heard that was a good range and leaves lots of headroom. I use Ind. Monitoring level to crank it up so I can record at this level, otherwise it's too quiet to hear over the beat. I like to perform with lots of volume. These settings seem to work well, and NO i don't have a noise problem. Actually after alll the compression and everything was said and done. The "silent" parts between phrases or words were actually pretty much dead silent. So that seems to tell me my gainstaging is correct?
Also, the multipressor would address my Highs in my voice jumping out at times. If I could just compress the Highs and not so much the lower frequencies. That might give me better results than JUST compression the whole spectrum equally.
Maybe, but I still think it's overcomplicating things, although it's hard to say without knowing the recordings. if you have particular sections which are harsh, the common technique employed by the big guys is to split the vocal ("mult") into multiple channels for the appropriate parts, so you can EQ and treat the parts separately. Some producers will even go so far as to automate or mult individual syllables in a vocal phrase, although that is extreme.
yeah, haha, It already took something like 4 hours last night to do what I did. I broke down and manually automated nearly every phrase/syllable in the song!!! Boy, that was work. But, the results were quite apparent. Even taming the out of control stuff, it gave life to the stuff that used to get kind of lost in the mix. I been reading the manual to get some easier shortcuts and stuff for editing automation, so that should cut some time a bit. Also, I'm not gonna be recording much more backing tracks I don't think. If I do, I will probably just copy the automation from the first "Lead" track to the backing track. I have gotten a little more comfortable using reverb to help fatten up my vocals.
So I used manual automation edits. Rider Plugin. Corrective EQing. Multipressor. "Color" Eqing. And Finally SilverVerb.
Sounded much better than before. And really I only spent about an hour or 2 more than what I had originally spent battling with plugins that couldn't correct a bad input. So good in good out and you know the rest. And like I said. I accidentally erased most of the automation I did, then saved then quit. Didn't realize I did that. Oh well. I'm learning