Compressors and vocals question

famous beagle

famous beagle

Well-known member
I'm just wondering how people approach this. By a show of hands, how many people typically compress a vocal (with outboard or plug) while recording?

I was just wondering because I was recording my wife singing a version of U2's "With or Without You" tonight, and the level varies dramatically from the verse (soft and gentle) to the "chorus" (I guess ... the "I can't livvvvvvvvvvvvvvvvvvve ... with or without you.")

I don't remember exactly, but I think it was probably coming in around -30 or -35 so on the quiet parts and it hit -12 I think on the loud parts, but I had a compressor (Art Pro VLA II) going for those louder parts. It ended up squashing more than I wanted on those parts. So I have no idea what it would have hit without the compressor, but I think it was doing about -10dB of GR or more at times.

We really liked the verses, so we're going to retrack the loud parts and set the level independently. How commonly do y'all do that? I mean ... set levels for quiet parts and then reset the levels for louder parts?

The problem with doing that, obviously, is that you can interrupt the flow of someone who sings well (like my wife), because she's capable (unlike me) of getting a keeper in one continuous take.

So ... I know I've asked several smaller questions throughout. I guess the big question is this: how realistic is it to be able to record any vocal in one continuous take without changing the level? Do engineers ride the faders (or preamp gain) more than I realize they do while recording a vocal?

I know I've rambled a bit, but I hope you understand what I'm asking.

Thanks
 
Some thoughts from me:

1 I don't use any compression on the way in. There is nothing between mike and interface except the mike cable.

2 I don't ride the (in my case) the pre-amp gain. It is set at a level, and that level is one I use for nearly everyone.

3 I usually get the person to sing through the song while I get myself organised as a warm up for them. However, I use this to judge what is going to happen.

4 If I hear that there will be variation in vocal level, I may modify the gain to cope with the loud bits (and let the quiet bits sort themselves out). Alternatively, I may just ask them to step back from the mike on the louder bits.

5 I have this aversion to messing with the sound on the way in . . . for the very reason you've discovered . . . once you recorded a modified sound, you can't unmodify it, and your plan B is to re-track.
 
I never use compressors on the way in. I set the gain to one level to handle the full range of dynamics. And I ride my distance from the mic while singing, which is proper live mic technique. I get in closer during quiet parts, and get further away during belts (sometimes 2 feet). If I find that the recording has too wild a swing in range, I will track the soft parts and loud parts separately. This works well when the soft and loud parts are different sections (like soft verse, then loud chorus). When there's a large dynamic range within a verse, or within a phrase/sentence, then you need to ride the distance from the mic.
 
How is her mic technique? What is her skill level regarding moving into and away from the mic.

I'm just asking because I've recorded many a good singer and have only had to very slightly ride the fader when recording, and have only had to use slight compression upon mixdown.

Edit: reading all the posts, I see Brian brought this up as well.
 
Thanks for the thoughts y'all. Her mic technique is pretty good, and I be sure to let her know to back off a bit when she got to the loud part. The problem with that (to me) is that if you back off too far, the sound really changes, IMHO.

I guess I could have set the pre really low to handle everything and I can try that the next time I have a song with a really large dynamic range. But ... I don't know. It seems as though if I had set the pre level to where it would have come into Reaper around -15 or -18dB or so (which is where I'm told you don't generally want to exceed), then I don't think she would have been able to hear herself in the phones at all during the quiet verses.
 
Of course I'm just used to an analog console and tape. But I'd imagine the techniques would transfer to a daw as well.
Get the max level that the singer would use in the song, set the trim and fader to that. You know where the peak is now.
Get them to ride it themselves with vocal technique. With a decent set of cans so they can properly hear themselves, the singer has more control over dynamics as well as peaks

For myself I've never noticed a tone change when they back away from the mic.
At the same time, I'm on that fader the whole time should I need to adjust it.
In the end there is very little compression needed. (Usually)

But that's just one technique I've used and has been my experience with analog and some of the singers I've worked with.

Edit: you can ride the monitor mix for them also as a method of controlling levels. Bring it up, the singer will bring it back. Bring it down they will sing a bit louder. I know it's trickery , but its a way of producing a result you want.
 
Last edited:
I use that same compressor, and I put it in an insert loop from my preamp. I'm not of the "but I can't change it later!" school of thought. I found a setting that works well for me so I'm not all that worried about it. I don't have a very dynamic voice, so my loud and not-so-loud passages aren't as disparate as some singers. I back away when I'm singing louder, and get close when singing quieter. I don't get closer than about 6" or else proximity effect is too much, and I don't back away more than about 18" or 24" because room reflections get to be too much.

I'm pretty shameless when it comes to compression and gain reduction on my vocals now that I'm using the Pro VLA. I use about an 8:1 ratio and shoot for about 8 dB of gain reduction. That sounds really aggressive on paper, but it works for me. I'm happy with the results. But again, I don't have a big powerful voice so maybe the rules are different. Just sharing what works for me since we're using the same compressor.
 
Just got one of those vla comps Still learning it but gotta say not bad. Glad it does what you want it to do for you. :-)
 
Tadpui said:
I use that same compressor, and I put it in an insert loop from my preamp. I'm not of the "but I can't change it later!" school of thought. I found a setting that works well for me so I'm not all that worried about it. I don't have a very dynamic voice, so my loud and not-so-loud passages aren't as disparate as some singers. I back away when I'm singing louder, and get close when singing quieter. I don't get closer than about 6" or else proximity effect is too much, and I don't back away more than about 18" or 24" because room reflections get to be too much.

I'm pretty shameless when it comes to compression and gain reduction on my vocals now that I'm using the Pro VLA. I use about an 8:1 ratio and shoot for about 8 dB of gain reduction. That sounds really aggressive on paper, but it works for me. I'm happy with the results. But again, I don't have a big powerful voice so maybe the rules are different. Just sharing what works for me since we're using the same compressor.
Click to expand...

If you are doing a lot of your own stuff and you know what to expect, then this approach makes perfect sense. When I do my own stuff, I have 'go to' settings that don't change, so it really makes no difference whether they occur inside or outside the box.
 
Well, the human voice is maybe the most dynamic source of recording ever. It is completely expected to get these amplitude variations you've got and I'm shure you know that.

Now, I like what RFR said, about having the singer doing the loudest parts of the song for level monitoring. Once you've set the maximum peak level (I tend to like having it -7dB or so), you know that at least you are not going to be clipping.

And about using the compressor, I think its a good idea. I like outboard equipment a lot, I think it gives a nice touch to the recordings, way more than plug-ins and stuff.
 
i always use compressors on the way in.

i use them for color, mostly, and some mild level control.

it is critical to spend as much time as is required, to get the levels set just right.

not a place to be in a hurry, or NOT understand the source fully, or to set thresholds to aggressively.
 
I also sometimes use compressors on the way in, but there are pitfalls. For example, if the singer is hearing the compressed sound they might not control their own dynamics as well.
 
I don't actually print compression on the way in. In fact, I don't really remember if I even own a hardware comp anymore. Did I sell that dbx, or is it in a closet somewhere? I do monitor through compression, but that gets replaced at mix time. I just set the gain so the mic wont clip, try to work distance, and then do whatever I need after. There is something to be said for committing to a sound at tracking, but it doesn't really work for the way I usually do things, and frankly the issue in the OP is part of the reason.

I find it easier to get a more consistent sound from a vocal if it is level led before it hits any transient shaping or "color" compression. That way each phrase hits the comp about the same and...whatever...it works better for me most of the time.

There was a trendy technique went around a couple years ago that might work in this type of situation, though, especially if you actually kind of want those dynamic swings between like quiet verse and loud chorus. Was called "Two stage compression" and was basically a parallel compressor into a series compressor. The parallel comp was set to shape the quieter section, but then when it gets loud enough that the parallel mix didn't make much difference, the series compressor handled shaping those transients. I've never done it, and most of the demonstrations I saw were on a drum bus, but theoretically should work for vox too.
 
I always compress on the way in, I use the compressor in my Langevin DVC. It's an LA3A style compressor. It can go really deep into gain reduction without sounding squashed. Since I do a lot of metal that goes from whispers to screams, it was the best way.

One answer is mic technique. When she winds up for the really loud parts, she needs to back off the mic to compensate for the volume jump. When doing the softer parts, lean into the mic.

Another way to do it is to set the levels so she can't clip without a compressor, then adjust the levels of the performance in the daw to fit the song.

Or, any combination of all three will get the job done without the vocal getting squashed.
 
I'm confident enough with a compressor that if/when I have a nice one (like a real 1176 or distressor) I would absolutely compress on the way in.

That said I don't have anything that nice in my home studio so no I don't print record with a compressor on the way in. I do with bass guitar though.
 
The problem with working the mic, as stated somewhere above, is that you can go in and out of proximity effect. If you have one and it works on the voice being recorded an RE20 or other Variable-D EV mic is a good option if the singer wants to work the mic.
 
GONZO-X said:
i use them for color, mostly, and some mild level control.
Click to expand...

I often find myself compressing vocals on the way in for tonal reasons. The signal will barely hit the compressor, but simply passing the vocal through the unit can add a very sweet character! I don't think this is essential by any means, but is very fun to play around with and if it sounds good, use it :)
 
You must log in or register to reply here.
Back
Top