Vocal Compression

[cdavis6406]

Im recording a session where the vocals alternat between loud(Scream) and a more subtle singing. The mic technique (distance $ etc) are ok. But the db differential between the to is still pretty extreme. When i compress these the loud parts and sounding cliped and static like. Do I need to go through the track and normalize the differnet sections so the volumes are comparable, then compress.
Or is there another more standard technique.

[M.Brane]

You could use plug-in automation to change the settings for the different parts or split the parts onto different tracks whichever works better for you.

I wouldn't normalize though. Leave yourself some headroom. The mix will thank you.

[cdavis6406]

Thanks man,
if I separte into track shouldnt i use trim to bring the loud parts down before I compress?

[albee1952]

That is sort of what I was going to suggest. You could also use the AudioSuite gain plugin. Depending on how strong the signals are at each extreme, I might add gain to the soft parts instead of lowering the loud parts.

[Petander]

Parallel compression added to the above may help,too.

Duplicate the track and use compressor to squeeze the heck out of the other one, EQ to taste.

Skycraft posted this on another thread: http://www.recordinginstitute.com/R2KREQ/excomp.htm

[BryanW]

First of all, in the future if you know that you're going to have a significantly louder section and you don't mind doing punch-in's, you could record all of the quieter sections with more gain on your preamps, and then you could go back and record all of the louder sections with less gain. You still might want the quiet sections a little quieter, but you don't need to have them at such extremes.

Next, there is the option that a few people mentioned of using the AS gain plugin to bring the levels closer together.

Recently, I picked up a great technique that's used by a lot of the big names in mixing. To my ears, it makes the vocals a lot more intelligible, and retains a very musical performance. First, go through with the gain plugin and even things out a bit. Play around with different gain increases until both sections seem to sit well. Next, you should really spend a good deal of time on riding the fader through your vocal takes. Make a pass through the whole song, and then go back phrase by phrase and syllable by syllable ensuring that every word is clear and even sounding. You might have to draw in some bumps for individual notes or syllables. Don't be afraid to have the fader moving quite quickly and quite frequently. Already you should be noticing a big improvement in your sound, even with no compression or eq. The vocal should seem to gel a lot better with everything else. Next, buss your vocal track to it's own aux, and put your compressor, eq, deesser, etc. on that aux. That way, all of those fader moves will control how much signal gets to the compressor. You can use a pretty low ratio and high threshold and still have a very clear sound. If you want additional compression for stylistic reasons, you can still add it here, but it wont have to work so hard if it's fed by a track with a good fader ride. Finally, set up inserts to additional aux tracks for your FX (delay, 'verb). Because the vocal aux is acting more like regular audio track now, you don't want to stick the FX on that track.

Hope this helps. Let me know if you have any questions.

-bryan w

[roberto_alayon]

When I record loud voices I usually put there some reverb an mix it down and put mid way to the rigth or left, it makes the scream to be just a great effect whitout loosing the force in the voice.

[roberto_alayon]

When I record loud voices I usually put there some reverb an mix it down and put mid way to the rigth or left, it makes the scream to be just a great effect whitout loosing the force in the voice.

[Mr.BossPlaya]

ride the vox fader then compress

[IntelDoc]

First of all, in the future if you know that you're going to have a significantly louder section and you don't mind doing punch-in's, you could record all of the quieter sections with more gain on your preamps, and then you could go back and record all of the louder sections with less gain. You still might want the quiet sections a little quieter, but you don't need to have them at such extremes.

I highly recommend this. Use different tracks too. Hell let him sing it all the way through, cut it out, add a new track and have him punch it all in.

USE a different mic for the scream parts. (SM7B works great for ROCK vocals in general, but fantastic for screamo's) I find a LD condenser is NOT the mic to go to for screamers but a dynamic mic handles the SPL's better and does not break up as much. That way you can also get levels dialed in better to start.

If you can, retrack and rethink it this way.

Then add compression.

Doc