ratiug wrote:
I saw this ( see below) info on another site which works well for backvox, but for adding compression he says " compress the neck out of it ( 10 db gain reduction) . Now, " compress the neck out of it " is a rather broad term. Can anyone suggest some more specific starting points , such as threshold, ratio, attack, release, knee ,output gain etc. That would work for this situation? Thanks
Here's a trick a learn recently for airy back vocals.
Send all your Back vox to a Bus called Airy ( per say )
On that Bus put ( let say ) The VC-64 in the Fx bin.
Hi-Pass everything under 4 or 5 khz ( to taste )
Add a hi shelf at 10-12K ( to taste ) and boost tremendusly ( 10db)
Then compress the heck out of it ( 10db of gr )
Blend with the original back vox... you'll get a beautiful shine and a smooth top end
Let's first sort out what he is doing.
He says send all you back vox as a group to a bus, for compression as a group. >>>>
The AW has no sub-group (AUX) buses that can be routed (internally) back to the stereo bus. So you would have to send them out via the AUX/Stereo outs, to an external compressor and then send that box's stereo return back in, through a pair inputs.. Or >>> you could make a sub mix, on the AW, of just the back vox tracks ( panned, eq'd compressed etc. using the AW's utilities) and then move this sub mix from its stereo track virtual back into the track section, to a pair of tracks and utilize this mix in your song. Following his advice, you would want this submix to be placed on previously un-utilized tracks, as you need to retain control of the original back vox tracks, for blending.
As for the mix you create to do this blending work - the quote suggests you do this with only the mid and upper mid sections of the frequencies. The quote advises filtering out every thing below 4 - 5 kHtz using a HighPassFilter, and also aggressively boosting the high frequencies at 10 - 12 K. so you are creating a "tinny" set of tracks in order to "compress the heck" out of them. As i made the sub mix I would probably do the EQ suggestions on each, individual track, so as to have some control of how much and what spectral sections of each vocalist's performance makes it to the sub mix, following generally the suggested filtering /boosting guidelines.
I would apply the compression to the group, as a whole, either on the stereo bus, during the sub-mix mixdown, or later, once the sub-mix has been moved back into the track section for blending.
Following the quote's suggestion would have you compress to achieve about 10dB of gr ( ie gain reduction - the meter bouncing downwards on your DYN screen)on this submix pair So you would adjust the threshold downward until the meter shows you this. Experiment with the attack/release times ( changes will be very subtle). I would probably use a knee of 4 or 5 and a quick attack (15 ms??) and slower release time 175 ms ??? but this is to taste, and as said the adjustments are often very subtle in effect. Make sure as the gain reduction gets more aggressive that you apply judicious amounts of makeup gain, in order that you get back something for all you are reducing. A-B comparisons are suspect if you are not in similar volume envelopes.
You are essentially creating a track of denser quality, in the upper mids, in order to hang it above the (uncompressed) back vox tracks that will run when you do your final mix down.
If you want to retain control of just how "dense" to make the submix, then do not apply the compression or any reverbs to the submix until it has been moved back into the song. ie. when making the submix, concentrate on balancing individual levels and colour, achieved through eq, pan and gentle compression of individual performances. I would avoid trying to make this mix "sweet" with any verb or EFF, as it may just sound fairly "???" following the suggested filtering settings. Its purpose in the final mix is to be more foundational than it is to sound sweet. Once it is back in the song, apply and tweek the compression and the verbs. It may not need much verb at all, if the original tracks are verbed up in the mix -- you choose how best to accomplish what your ears want to hear.
I would cautiously suggest that you may run into sibilant issues, should the bed tracks have sibilant qualities in the first place. It will depend on what SSSS SSHH and pops those tracks contain.
Hope this helps. I will try this suggestion myself when I again get into a project utilizing a back vox group. this technique may be useful with other groups tracks. Multi-mic'ed Acoustic guitars come to mind.