Compressing Vocals (from soft whispers to full on singing/screaming)

[JB872]

How do you guys compress vocals that change really fast from soft whispers (kinda like way Chino does from the Deftones) to full on singing/screaming? My guess here would be to double the track and use automation on both so the soft parts hit one compressor and the loud parts hit another, both with completely different settings. Am I close?

Right now I'm feeding the vocal through an LA2A -> 1176 (all 4 buttons in, needle pinned) for the heavy. Not too sure how to deal with the whispers.

Any suggestions?

[John Suitcase]

When I've done hardcore vocals, I've basically just compressed the hell out of them, using several stages to make the compression a little less obvious.

You can try moving the whisper stuff to another track, and using different processing on them, which might make the compression less obvious.

But yeah, I would compress pretty hard during tracking, then again during mixdown.

[JB872]

Sorry, I forgot to mention that the vocals were tracked by someone else without any compression.

[turk sanchez]

I like tracking with an 1176 or a Distressor ...but in your case I would go with this... If it is soft in the verses and loud in the chorus split the chorus to a new track. I might use the same compressor settings on both inserts but just change threshold/make up. Depending on what you have I would use something like a Distressor, 1176 or an 1176 plug in, Smack etc...fast attack and medium to fast release...4:1 or more with about -10 to -15 on the loud stuff (to tase...but slammed). Even more if using a Distressor...even -20 db! I will let the nedle on the 1176 go all the way back sometimes...but I don't tend to dig high ratios on vocals for some reason.

If the singer is all over the place I might leave it all on one track. Maybe follow it with a peak limiter (elop, la2a etc) and just have it bump here and there. Then ride the faders. If that can't get it I don't know what else could help...Good luck bro.

Waffle Waitress

[bringmewater]

Ride em like a rented donkey !

[joaquin]

Originally Posted by JB872

Right now I'm feeding the vocal through an LA2A -> 1176
Any suggestions?

I'd put the 1176 in front for the fast attack and the LA2A for the sustain.
and the faders .........

Originally Posted by bringmewater

Ride em like a rented donkey !

[JB872]

Great ideas so far. Keep them coming

[ZEUSS]

For the chino style vocals, compression alone isnt going to cut it. You gotta do some rides. Or like john suitcase mentioned you can also split the parts into a few tracks with some compression and some rides.

[eso]

For vocal performances with huge dynamic range, I usually split different sections of the vocals to different tracks (presuming you have enough) so that I can process them seperately/differently.
I also draw volume envelopes/fader automation before hitting the compressor (and after if required) and use more than one compressor in a chain. Using a multiband compressor into a broadband compressor, then into a limiter can allow for huge cumulative compression with less obvious artefacts.
Adding distortion also compresses well if it works within the track.

[turk sanchez]

I gotta take back what I said about not liking high ratios on vocals... I used a u195, great river, distressor in nuke for vocals tonight...sounded awesome...very nice I thought.

[7rojo7]

Split the whispers, singing and screaming or subsonic demonspeak to seperate tracks, if out of the box, mult and bring up on different channels and mute those you don't need when you don't need them, if in the box, make some new tracks and drag the desired parts to their tracks. You might want to copy the parts over so you can use parallel processing, leaving the original unprocessed and adding the other elements to it.
Ride the original so it sounds good but covered only over the bass and drums, copy the automation to the other tracks and add them to cut through the other stuff you also may want to duck the other stuff using the kik and the vox as the key.
I like riding the fader into the compressor, it takes up another channel in the DAW. Between the original, the splits and their processing channels, the FX and a main vocal sub you could have up to10 channels for LD VOX. Is that too many?
LD VOx, whisper split, 1up1dn whisper, sing split, sing dly, scream split, scream plate, LD VX SB
I like to hit the limiter at the mastering stage, I compress for musicality for the mix. The Sony inflater could work for the whispers