I'd go for the automation. I never use gate on vox since it's pretty easy to rubberband the automation just looking at the file.(assuming you are on a daw) If you are on a board, just do some mute runs. Gates are tricky with vox, especially if they are not the best ones. Also if you are gonna use a gate, put it before you compress.
my 2 cents

if you dont notice it in the song your ok with the gate

"Gates are tricky with vox, especially if they are not the best ones. Also if you are gonna use a gate, put it before you compress."

Yeah, I should have put down the EQ list.. Cubase SX, VST dynamics - Gate and Compressor plug ins. When you say gates can be tricky how so? It seems to me that if you set the db you want it to allow audio " out of the gate" so to speak, it cuts out anything below it. Is it whats is happening as the gate closes that could be a problem?

i would also say use automation, even quicker you can (if your using protools) use the strip silence comand to physically remove audio regions like the gate would, though its better cause you have more controll, after you cut the vocals into regions with strip, go back and listen to the track to find any funny areas which you can just pull out the "bad" edits. and if theres some really dynamics parts which are hard to get right with the gate you can strip the tracks little by little for more controll, kind of like automating the gates threshhold.

I'd be trying to use a good surgical eq or filter and see what I could do before I would even try a gate. As far as the compression, if adding it is making it that much worse, then don't use it. Despite what you may hear or read, you don't always have to use compression. You have to consider more than the exact sound/tone you like or are trying to achieve if you are doing more harm than good to the overall quality of the source. Mabe next time the person that recorded it will learn to pay more attention up front while tracking to help avoid the kind of problems your dealing with now. Do what any real engineer would do. Try every possible technique you can come up with and pick the one that does the best job with the least amount of damage. This would be a good time to explore the use of the sidechain function of a compressor like you would as a de-sser but with the focus on the frequencies of the problem areas.

Hey,

Yea, I would definitly go with placing the noise gate BEFORE the compressor, if you decide to use it. But I would probibly attack the problem first hand with eq. The 'thumps' are likely to be of a low frequency not present in the VOX, so cut them out. If your not using a 'multiband' compressor, them these bumps would have an influence on the compression. You could also manual edit out the bumps as well.

"Gates are tricky with vox, especially if they are not the best ones."

I think what he meant was that since the artist wasn't very good at "working the mike" this will make setting your rise time of the gate for the initial scream hard to adjusted. You need to set how fast the gate is opening and how slow it is closing.

Sounds like your singer is inconsistant. I say use the fader with automation, and set the compressor as a compliment for the suddle variations. Now you shouldn't really need a gate, the volume automation will sound more natural.

Also i'm curoius to know what else you hear on the track besides his breathing and foot? Are you only noticing these impurities with a solo in place? If so then it's not such a big deal. But if your song calls for a solo then I think the fader and compressor should cure his breathing problems. Clicking and tapping will be more of a challenge. Once again I don't think you should use the gate because even a fast fade in/out is much more natural sounding. Practice drawing automation manually, it can be a fast tool once you learn it.

AudioGaff you mentioned the use of a sidechain. I'm familiar with using sidechain circuits; mostly in regards to mixing drums, but what are you suggesting he sidechain to the compressor for vocals? Or are you suggesting he sidechain the compressor to something? I'm curious to know more about this. Could you explain a little more?
Jeremy doesn't have sidechain capabilities but i'm curious to know anyway.

Originally posted by Jeremy Sicely:
...stomping a bit on the floor with his foot through out the track, needless to say when I compress for the vocal sound I want these levels are brought up. This is only real noticable when the track is solo.

First in the signal path, use a high pass filter between 100 and 200 to remove much of the low frequency thump from the foot. Then compress to taste. If it "is only real[ly] noticable when the track is [in] solo."...then the above will help.

Originally posted by Jeremy Sicely:
The singer lets out a couple of heavy breaths

If you have automation just ride those down. If you do not have automation, bounce them to another track and back again at a lower level. Try around -10db as a starting point.

Well, I am not really a recording expert, but I record vocals a lot.
Since your singer has his own track, which by definition can be soloed, why don't you simply cut the breathing parts ??? and thereafter you merge what is left... it's easy with DP, (you can even get away with a sneeze, a cough or whatever...) and it must be so with Cubase...
As for the thomping, I think like many others here that EQ is the good way...
Good Luck

Peter

AudioGaff you mentioned the use of a sidechain. I'm familiar with using sidechain circuits; mostly in regards to mixing drums, but what are you suggesting he sidechain to the compressor for vocals?

Much like the way a de-esser works and the way you use a compressor and an eq in the sidechain to make your own de-esser, you use the eq to make the compressor more sensitive specificly to the problem freqs so that it reacts as a ducker. If this works for you, then you can further process and may even be able to add another comp in the chain to get the tone/sound you wanted in the first place. Using the side chain of a compressor is an under used but powerful way to control and shape dymanics.

Yeah I put a low end filter on and automated the silent sections. This is by far much more, as AudioGaff said, "surgical" and precise. I always first EQ and then compress to help balance the level.