Vocal recording + clipping woes.

[Horkin My Lunch]

Yesterday, I was recording a band, and everything was going smoothly until it came to vocals. The vocalist has a range of soft passages to shouts (I say "shouts" because it's not exactly a scream, but it's fairly close) and I was having a hell of a time trying to bring that in without clipping. Granted, I was using a Rode NTK, which is a retardedly sensitive microphone, but I was wondering if there was anything you guys do to keep the clipping in check while making sure the signal is optimal for the quieter passages in a single take. I guess I could use a limiter, but I've always been wary of recording with a plugin on. Furthermore, I'm also curious to see how you go about leveling the vocals when tracking (getting the signal as hot as possible for the vocalist was tough).

 

[BushmasterM4]

Get the vocalist to redo if possible. Have him back away from the mic when he screams or gets louder. If you cant redo the vocals, then slice up the track and get the vocals sitting close to what you want by adjusting each slice or clip's volume. Once that is done then compress the hell out of it.

 

[iqi616]

Furthermore, I'm also curious to see how you go about leveling the vocals when tracking (getting the signal as hot as possible for the vocalist was tough).

I use a level where the loudest section is well clear of clipping and I just let the quiet sections look after themselves. At mixdown I'd either use automation or shift the loud sections out to a spare track.

Don't fuss about recording hot - modern gear is so clean (especially in 24-bit). Recording hot is a throwback to narrow-track analogue and early 16-bit digital multitrack and it just makes you obsess about clipping instead of concentrating on more important things.

 

[legionserial]

I would say that if you can retrack, keep the loudest parts well below clipping (you may need the headroom for later on in mixing), and don't worry too much about any weakness on the signal in the quieter parts. Digital has a pretty low signal to noise ratio in itself an so will lend itself better to signals at the quiter end without too much noise, provided the SNR in your room isn't a problem when recording that quietly. If the performance really is too dynamic for your liking, you could compress a little on the way on I suppose, but that wouldn't neccessarily be my preferred way of doing it because obviously you can't undo any compression you used on the way in. You can always compress the hell out of it or use volume automation in the mix as you see fit.

What sort of dbVU would you consider to be non-optimal on the quieter end?

 

[RD423]

I wouldn't worry about it clipping unless it starts distorting "clicking/popping & all that kind of stuff" in mixdown I would chop it up have the quiet parts on one track and the louder vocals on another.

 

[Bguzaldo]

I usually just ask them, are you gonna yell or sing? And then I say ok, well were going to do the yelling parts now(then i check my levels, adjust accordingly, record the takes of the yells only), then do the singing softer parts on a separate track(again re-check, re-adjust, and then record) and then i slice them up nicely and apply the same eq's to both and compress them so they seem to fit together nicely.

Works for me, i dunno if most singers like to go thru it live all the way, but most guys i get don't mind

-Barrett

 

[VomitHatSteve]

I usually just ask them, are you gonna yell or sing? And then I say ok, well were going to do the yelling parts now(then i check my levels, adjust accordingly, record the takes of the yells only), then do the singing softer parts on a separate track(again re-check, re-adjust, and then record) and then i slice them up nicely and apply the same eq's to both and compress them so they seem to fit together nicely.

Works for me, i dunno if most singers like to go thru it live all the way, but most guys i get don't mind

-Barrett

+1 for this process, although I'd recommend doing the singing parts first. (Why thrash your voice with yelling before the singing parts?)

 

[LeeRosario]

I personally try to go tube on mics whenever possible on the vocal session. That and establishing some type of optimal compression on the way in.


For me personally, the mentality has always been to anticipate the mixing process by what I do in the tracking process. Hence, if I compress the vocals as they are laid down, I'm not going overboard with the compression settings; just enough to keep it in a level realm.

Either way, depending on how dense the mix is, you're going to find that heavy (yet proper) compression and limiting can be your saving grace during the *mix phase*. In the words of mixer Chris-Lord Alge, it's like "putting a leather jacket" on your singer by giving the vocals attitude.

Enough compression during tracking to keep it controlled. You can't undo that mistake later on if it comes out too aggresive.