Compression on vocals, before or after recording? Or not at all?

[gfsark]

Just attended a recording session at my school. The instrumental had already been laid down, so the exercise was to overdub the solo vocal track. The prof said that he always runs vocals in this situation through a compressor before recording, at a compression of about 2.5 .

On one take I was in the studio and the other take in the control room, so I got to hear the effect of compression. To my untrained ears, compression made the voice seem much more like a recording, and definitely took an edge off the live performance. Maybe that's the price you have to pay to get a balanced recording?

I'm curious about how you use compression on vocals? Thanks

[Bob Womack]

I typically don't compress during the recording phase except when I've got an unreasonably dynamic talent. A trick I learned from the film automatic dialog replacement community is to run the mic into two channels. Set up one channel as the main track with no compression. Set up the other one with either 6db less gain or a high-level limiter to allow you to catch "overs." That way a take everyone loves isn't lost if the talent throws out an unexpected excursion.

But dynamic control is a fact of life if the person is singing over a combo. I just wait for after the performance to apply it. That's the beauty of modern equipment with low noise floors. When I get into the mix prep stage I will use a combination of gentle off-line compression and generous hand-written automation to bring the performance into a dynamic range that allows it to sit well over the band.

And here is the fact of life: A combo environment is an unnatural environment for a voice. We are used to free vocal expression using pitch and dynamics to get our point across and the combo imposes what amounts to a noise floor under which the voice can't duck without being lost. As a vocalista becomes most experienced, he or she understands that and projects more. However, they still often need level help.

Here's another fact of life: You may have never heard a piece of music broadcast without dynamic control. Radio stations know that the music will disappear into the background of a car or subway or office or whatever if it is allowed to sag to a low level. That gives you the choice of leaving the dynamic control to others downstream or making the decisions and artfully controlling yourself.

Bob

[muscmp]

great info bob--as usual!

play music!

[gfsark]

Yes, a great explanation! Thanks.

[DupleMeter]

Personally, I like to have a small amount of compression on the way in for a VOX and then add a small amount during the mixing. To my ears it's much more natural than a single, heavier compressor (either during tracking or mixing).

Everyone develops their own tactics, so try everything and find what works for you...and then resign yourself to the fact that nothing works 100% of the time.

[KevWind]

Yes I use compression, but so far have not used it while tracking.
As Bob mentioned it would be extremely unusual to hear any commercial recorded vocal without dynamics control and that is often also true for professional live performances.

Now I am not certain I understand exactly what it is you mean by "took an edge off the live performance"

Because whether live or recorded (if compression is used correctly) it will actually enhance things like presence, air, intimacy and depth ..... if used incorrectly can begin to detract from those things .


But the two channel thing ( 1..with comp and 1.. dry) Bob was talking about is something I may experiment with a bit , now that I have a very nice tube hardware compressor to work with.

Also just to expand a bit on DupleMeters comment it is also possible to actually do two compressor passes during mixing and achieve much same type effect as doing one during tracking and one during mixing.

I also use a ELOP limiter on my vocals for live performance specifically for a bit more forward presence

[Ty Ford]

Depends on the talent and the gain reduction device.

Yes, I use a compressor while recording most vocals, but it's a very good compressor and doesn't generate unpleasant artifacts, especially the way I have it set.

I may use more in mixing if I think the vocal needs to be more persistent to prevail in the mix.

About the radio thing. Sort of. I was in radio for 17 years here in Baltimore and down in Washington, D. C.. Radio stations and TV stations use compression and limiting in the audio chain, not so much to keep the road noise from eating up the sound, but as a competitive strategy to sound louder than the competition.

Classical Music Stations usually use less than rock stations and that's pretty apparent when you go up and down the dial.

The other thing all that mediocre gain reduction causes is making inhales sound asthmatic because the circuitry raises the level of quieter sounds.

If you can, bring the signal into something with meters on it and you can see the altered dynamic range. On some stations, the meters barely move.

Regards,

Ty Ford

[Mobilemike]

It depends on your system. If you have a nice hardware compressor that you like the tone and vibe of then yes, compress on the way in. You will get closer to the final mixed sound right there at the recording stage. Just don't go too heavy handed on the compression as you can't undo it once it's there!

If you have software compressors only though, no hardware, there is nothing to be gained IMO by compressing on the way in - just do it during the mix stage.

-Mike

[MikeBmusic]

I only record myself, so I have to 'work with what I've got'. I never record with compression on. During the mixing stage I will automate volume as needed to keep it at the right level in relation to the instruments. Then I'll add slight compression after.

[Wyllys]

Originally Posted by gfsark

Just attended a recording session at my school. The instrumental had already been laid down, so the exercise was to overdub the solo vocal track. The prof said that he always runs vocals in this situation through a compressor before recording, at a compression of about 2.5 .

On one take I was in the studio and the other take in the control room, so I got to hear the effect of compression. To my untrained ears, compression made the voice seem much more like a recording, and definitely took an edge off the live performance. Maybe that's the price you have to pay to get a balanced recording?

I'm curious about how you use compression on vocals? Thanks

To put it bluntly, if your instructor actually tells you that he "always" does something, be suspicious. Situations differ and any decent teacher will give you the information and techniques required to find the proper solution du jour.

If there is a standard industry rule for tracking it would be to track dry, the reason being that once "in the can", such applications cannot be undone and you're stuck with something designed for the track itself and not necessarily for the over-all mix.

In the case you describe your instructor may be said to be wrong...very wrong.
He must be one of the new breed of techs that compress everything at every stage and end up with ZERO dynamic range and the now common "mega-squashed" product.

[Bob Womack]

By the way, saying I don't use compression while recording doesn't mean I don't have a compressor inline. Distortion is simply a no-no in professional recording. It is fascinating that the higher pressure and speed a session features, the less forgiving the client usually is of mistakes and flaws, including distortion. The two main types of preamps I have in my suite, Avalon VT77s and Neve Portico 5015 pre/comps both feature excellent onboard compressors. I usually put them inline and set them with a gentle ratio and a high-enough threshold that they are rarely utilized. I might see an occasional two or three db of compression on an over. I gently ride the gain during the session to make sure the compressor rarely gets hit.

Bob

[Ty Ford]

Like Bob,

I use a Millennia Media STT-1. It's more like a channel strip with EQ and gain reduction options. I have it at low settings and keep an eye on the amount of gain reduction, but catch the occasional peaks at 2-3 dB with fairly slow attack and release times and around 2:1 ratio.

I save the heavy lifting for the plugins during mixing.

Regards,

Ty Ford

[KevWind]

Originally Posted by gfsark

Just attended a recording session at my school. The instrumental had already been laid down, so the exercise was to overdub the solo vocal track. The prof said that he always runs vocals in this situation through a compressor before recording, at a compression of about 2.5 .

On one take I was in the studio and the other take in the control room, so I got to hear the effect of compression. To my untrained ears, compression made the voice seem much more like a recording, and definitely took an edge off the live performance. Maybe that's the price you have to pay to get a balanced recording?

I'm curious about how you use compression on vocals? Thanks

Don't listen to any totally subjective declaratives from any direction .
However to clarify, many professional recording engineers do routinely use compression in tracking vocals. (no doubt for many of the same reasons Bob mentions above). And many of these same engineers produce some of the best mixes made, and with great dynamics. So your prof in stating his preference was not wrong .
Also there is nothing new about compressing while tracking , it has been done for as long as analog compression has been around

There is no more an industry standard to record dry as there is to compress, both are done routinely. It completely depends on the engineer's personal preference period .

Also note as a idea to place into your learning curve, it is pretty much impossible to accurately compare what you heard "out in the studio " with what you heard " in the control room" as some indication of the effect of compression.

Here is an example of a vocal tracked with a compressor in the chain .
The tracking chain, "For the track Don’t Know Why, Newland had tried using a Neumann U47 and an AKG C12vr, but eventually chose a Neumann M49 to keep the full natural warm tones. This mic went into a Manley tube pre for added warmth, and then into one side of a Manley Vari-mu compressor, turned to a fast setting for both attack and release. The threshold was held high so that it rarely touched the compressor. "






[Bob Womack]

Mmmm... M49. My first two years as a recording engineer I recorded classical performances at a facility that owned a pair of Neumann M49s. That was 1979-81. Wow, what a way to start a career. The M49 is a gorgeous mic. In 1983 my next company acquired a '57 Telefunken U47 with eight hours on it. In the '50s Telefunken distributed Neumann in the USA and rebranded the products. I sent the mic to then-distributors for Neumann, Gotham Audio, for a refurb. It returned with all the plates replaced to rebrand it back to Neumann. Boy, was I hacked off. But it was, and is, a great mic and now lives in the studio I am assigned to.

Bob

[Wyllys]

Originally Posted by Bob Womack

I gently ride the gain during the session to make sure the compressor rarely gets hit.

Bob

That's called "digital" compression: using your digits on the faders.

OPs instructor was cited as routinely using a 2.5:1 ratio with no mention of threshold. I'll admit to using a 1.2:1 or 1.4:1 ratio in some cases, but as you said a fairly high threshold. And I prefer using a multi-band comp with "look ahead" circuitry. Like you, I usually see no more than 2-3dB gain reduction...if that.

I think the point of confusion for beginners is not understanding the use of a good compressor as a limiter to avoid overs rather than an "omni-squash" and killing the dynamic range.