Can anyone give me some tips for compressing a very dynamic vocal ? So far I have steered away from recording certain types of song, simply because i'm not sure how best to record a vocal that goes from a whisper to a shout ?

--------------------

www.myspace.com/joekmurphy

It depends on how wide the source range actually is. In many cases some moderate compression on the way in can help, provided you are able to find a good setting that will suit the finished track, and you can then polish the balance and dynamic control within the DAW, either through clip editing or level automation, and/or more track compression/limiting. In the 'olden days' producers or engineers would 'ride the fader' on the mic channel feeding the tape recorder -- applying manual fader automation to control levels through anticipation... a long lost skill in many parts these days!

In extreme cases, I find the easier solution is either split the mic to feed two channels of preamp with radically different gain settings, or rig two mics feeding two preamps, again with different gain settings. Set one channel to cope well with the quiet bits (knowing it will clip in the louder sections), and set the other to deal with the loud bits (knowing it will be too noisy to be usable in the quieter bits). Record both outputs and then simply edit between them to build the raw track, applying fader automation, clip editing and channel dynamic processing as required to smooth it all out.

Hugh

--------------------
Technical Editor, Sound On Sound

Recording a dynamic vocal is just about following the usual principles of leaving plenty of headroom so that the peaks go nowhere near clipping - recording at 24-bit I'd aim for peaks of -10dBFS.

If the vocal really is so very dynamic that you feel the need to do something special, then you could simply track the quiet and loud parts separately, applying the same peak levels as above. This can often help the vocalist to focus, but it can also make the performance somewhat disjointed - it really depends on the individual. If you have to track the vocal in one take to get the performance then you can mult the vocal into loud/soft tracks and process each accordingly.

For me it's always about performance over technical perfection. With modern recording systems, as long as you have a reasonably tracked (and un-clipped) performance you can usually sort out the levels later.

--------------------
An Eagle for an Emperor, A Kestrel for a Knave.

Quote Hugh Robjohns:

---

...

In extreme cases, I find the easier solution is either split the mic to feed two channels of preamp with radically different gain settings, ...
Hugh

---

Simply using something like this, Hugh?

http://www.bing.com/images/search?q=microphone+splitter&view=detail&am p;id=765B57204A0B04E1CD9A87F0DAE4D01932BFC58E&first=0

Thanks!

--------------------
Alexis -Cubase 6.5.0/SX3.1.1.944, XP SP2, 4GB RAM (1GB not accessible, but used just to balance the computer so it doesn't tip over); Delta 66 in Omni i/O Studio; Motif8; UAD-1

Potentially, yes. I have used simple cable splitters like that before successfully. However, you have to be slightly careful about grounding paths to avoid ground loops and phantom power issues, and the mic will see roughly half the preamp input impedance which might cause tonal changes with some mics.

If you go down the cable split route one of the male XLRs has to be ungrounded (have an isolated pin 1), and phantom power can then only be provided via the other male XLR.

A transformer mic splitter is the other alternative solution.

hugh

--------------------
Technical Editor, Sound On Sound

Thanks for that, Hugh!

Quote Hugh Robjohns:

---

(Using simple cable splitters,) "the mic will see roughly half the preamp input impedance which might cause tonal changes with some mics."

---

Is there any way of predicting ahead of time whether a given mic will have a significant tonal change? I have an AT 4033a.

Quote Hugh Robjohns:

---

...
If you go down the cable split route one of the male XLRs has to be ungrounded (have an isolated pin 1), and phantom power can then only be provided via the other male XLR.

---

I wonder if these all come modified as you've described (pin 1 isolated in one of the male XLRs), or if I need to check this somehow before use. What is the catastrophe that will happen if used without that modification?

Quote Hugh Robjohns:

---

A transformer mic splitter is the other alternative solution.

---

I have found two models:

http://www.bing.com/shopping/engineering-js2-passive-microphone-splitter-d irect-box/p/9B2569D6B6BBC98481E2?q=transformer+microphone+splitter&lpq=transformer%20m icrophone%20splitter&FORM=HURE

http://www.bing.com/shopping/vtg-ss1001-mic-splitter/p/93879C52574FF163A43 D?q=transformer+microphone+splitter&lpq=transformer%20microphone%20splitter&FORM=H URE

The first costs $US 220. and the second one almost one-tenth as much ($35). What type of sound quality difference would one expect to hear when comparing the two units? For example, if one had an AT 4033a and used the splitter to feed the two preamp inputs of an [M-Audio Omni I/O feeding a Delta 66 card], would the difference in sound between the two transformer splitters be noticeable?

What about if the splitters were instead feeding the two preamps of an RME Fireface UCX - would the difference in sound between the expensive vs. cheap transformer mic splitter be noticeable then?

Just for reference, I found this article on microphone splitters: http://whirlwindusa.com/support/tech-articles/microphone-splitters/ .

Thanks Hugh again, and for any further answers you may have time to give (or anyone else!) -

Quote Scouser:

---

Can anyone give me some tips for compressing a very dynamic vocal ? So far I have steered away from recording certain types of song, simply because i'm not sure how best to record a vocal that goes from a whisper to a shout ?

---

Maybe this can help - http://www.soundonsound.com/sos/may11/articles/wave-rider.htm . The owner of Wave Rider has said on another forum that a Windows version is coming out soon.

--------------------
Alexis -Cubase 6.5.0/SX3.1.1.944, XP SP2, 4GB RAM (1GB not accessible, but used just to balance the computer so it doesn't tip over); Delta 66 in Omni i/O Studio; Motif8; UAD-1

Hi Joe,

I reckon there are two sides to this...

(1) When recording, try to keep the vocalist in a similar place and distance in relation to the mic. That way, when you perform any processing you won't be drawing more or less or different room sounds into the mix. That can otherwise stick out like a sore thumb and it's a b****r to work with when mixing. Nothing wrong with compressing on the way in if you want, but be careful not to screw it up. There's no need to do it then if you're using 24-bit recording.

(2) When mixing, if you're dealing with problems like this, try multing out the different sections so that you can apply different processing to, say, a gentle verse and a belting chorus. If using 'automatic' compression, then a couple of gentler passes can work better than one - depending what it is you're trying to do. Remember, though, that you can manage dynamic range with detailed editing and automation - you're not, um, limited to using compressors and limiters. Eg in Cubase, I'll often go in and edit the part, adjusting the envelope level for each clip, or will automate the gain control at the top of the mixer channel before insert processing. You then have the channel fader that you can also automate after any processing. Done well, such automation can be more effective than simple compression if you're trying to tackle specific problems. Of course, you can also automate compressor thresholds etc if you prefer to work that way.

Hope this helps.

Matt

--------------------
SOS Reviews Editor

Quote Matt Houghton:

---

Hi Joe,

I reckon there are two sides to this...

(1) When recording, try to keep the vocalist in a similar place and distance in relation to the mic. That way, when you perform any processing you won't be drawing more or less or different room sounds into the mix. That can otherwise stick out like a sore thumb and it's a b****r to work with when mixing. Nothing wrong with compressing on the way in if you want, but be careful not to screw it up. There's no need to do it then if you're using 24-bit recording.

(2) When mixing, if you're dealing with problems like this, try multing out the different sections so that you can apply different processing to, say, a gentle verse and a belting chorus. If using 'automatic' compression, then a couple of gentler passes can work better than one - depending what it is you're trying to do. Remember, though, that you can manage dynamic range with detailed editing and automation - you're not, um, limited to using compressors and limiters. Eg in Cubase, I'll often go in and edit the part, adjusting the envelope level for each clip, or will automate the gain control at the top of the mixer channel before insert processing. You then have the channel fader that you can also automate after any processing. Done well, such automation can be more effective than simple compression if you're trying to tackle specific problems. Of course, you can also automate compressor thresholds etc if you prefer to work that way.

Hope this helps.

Matt

---

Thanks for that also, Matt!

Tools like Wave Rider and Vocal Rider - in Cubase, would they be adjusting the levels before or after processing? If after, is there any way to have them automate the input trim instead?

Thanks -

--------------------
Alexis -Cubase 6.5.0/SX3.1.1.944, XP SP2, 4GB RAM (1GB not accessible, but used just to balance the computer so it doesn't tip over); Delta 66 in Omni i/O Studio; Motif8; UAD-1

Quote alexis:

---

Tools like Wave Rider and Vocal Rider - in Cubase, would they be adjusting the levels before or after processing? If after, is there any way to have them automate the input trim instead?

---

It's your choice, based on where you put them in the chain!

--------------------
An Eagle for an Emperor, A Kestrel for a Knave.

Yeah, you can use them wherever you want to put them. Personally, I like to keep the fader available for fine automation anyway, so I'd never use these in post-fader inserts (ie the last two on a Cubase channel). I'd also tend to use them after any processing that was going to change the level of the vocal, like compression and de-essing. But on the whole, good as these auto-level plugins are to a point, I'd much rather manually edit and automate levels. One pass to sort out the take, then do any processing to get a reasonable static mix balance, and then another to manage levels in the wider dynamic context of the mix. But that's a personal thing and YMMV...

--------------------
SOS Reviews Editor

Some great tips there..


Thanks to one and all

--------------------

www.myspace.com/joekmurphy

A well educated vocalist will know how to "work the mic" properly. I find that if I coach the vocalist a bit and have them move closer on the quiet and move back and turn a little off axis on the louder passages I need less compression. I usually point out that if the music gets buried by the vocal in the headphones, they are too close and if the music buries their vocal, they should move closer. They get better with practice.
Also, if I have already recorded vocals that are very dynamic, I may move the loud sections and quiet sections to different tracks and simply use separate compressors for each.
Cheers

--------------------
Jeff Ling
www.RECW.com

Quote Jeff Ling:

---

A well educated vocalist will know how to "work the mic" properly. I find that if I coach the vocalist a bit and have them move closer on the quiet and move back and turn a little off axis on the louder passages I need less compression. I usually point out that if the music gets buried by the vocal in the headphones, they are too close and if the music buries their vocal, they should move closer. They get better with practice.
Also, if I have already recorded vocals that are very dynamic, I may move the loud sections and quiet sections to different tracks and simply use separate compressors for each.
Cheers

---

I agree (as they say on a certain trashy u.s. tv show) 5000 %. I cannot believe its taken a whole thread to bring this up. Back in the tape days I always did this and helped educate vocalists about it.

Levels aside, a mic will sound Completely different when you shout into it at a certain distance versus projecting quietly at the same distance. By moving back/slightly off axis when you get loud it not only evens the sound and levels out but keeps your mic working at its best s/n spec.

Look at it this way - if you were far away and whispered the gain needed to make you audible would bring up much noise. And if you were close and shouted that might drive your mic or pre way too hard.

Get it right at the source. Then, a little nice LA or dbx squash is fine on the way to daw, just to smooth it out.

Just my 2NIS
MadManDan

--------------------
Gear list: If you can't find it, grind it

Yeah I'll second having an 'easy' compressor you can trust - an auto set and forget one. Drawmers' MX30 is nice and transparent, and a good DBX with an overeasy mode is useful. You dont have to spend a lot to get good results, but you do have to know how it behaves - then just forget about it knowing that it will happily sit there doing its job clamping down any excess without fuss.

--------------------

Quote alexis:

---

The owner of Wave Rider has said on another forum that a Windows version is coming out soon.

---

I'd buy Wave Rider like a shot if there was an AU Logic version ... Talk about one of the most useful plug-ins ever!

--------------------
PA stuff on FB

Quote:

---

A well educated vocalist will know how to "work the mic" properly. I find that if I coach the vocalist a bit and have them move closer on the quiet and move back and turn a little off axis on the louder passages I need less compression.

---

Although technique is vital, ive never had much luck with this approach, to me it always sounds like somebody standing close and then backing off. Maybe this works well when the dynamics are not so extreme.

--------------------

www.myspace.com/joekmurphy

Quote Scouser:

---

Quote:

---

A well educated vocalist will know how to "work the mic" properly. I find that if I coach the vocalist a bit and have them move closer on the quiet and move back and turn a little off axis on the louder passages I need less compression.

---

Although technique is vital, ive never had much luck with this approach, to me it always sounds like somebody standing close and then backing off. Maybe this works well when the dynamics are not so extreme.

---

It's wierd. Even though my education for inexperienced vocalists usually consisted of little more than "Stand here,ok. When you need to get louder, move back or aim away from the mic a little" I never had any real issues. I would record some practice takes to make sure they got it, and that the loudest parts didn't overload. Then just go. Maybe I didn't have a lot of people doing extremely quiet vocals.

That said, I can see there still needs to be a levelling method for after-the-fact, esp since there are people like me who in my current state have no outboard compressors at all.

I think a good idea would be to play the vocal track back, automate all the major level fixes, and buss that to a channel that gets compressed. This will keep everything evenly squashed.

Again, just my 2 NIS
Mmd

--------------------
Gear list: If you can't find it, grind it