join

The Proper Level For Vocals by fHumble fHingaz

Brandon Drury —  September 16, 2012 — Leave a comment

Man, Jeronimo!  Why don’t you ask us a really hard question!:rolleyes:

Seriously though, the simple answer is:  There is no simple answer! – it is ultimately sooo context-dependent.

the “big boys” mix in vocals at pretty high levels above the mix. When I do something similar, I usually get told the vocals are too loud.

This is usually a pretty good indication that the vocals need more compression/automation to level them out, or it could be that they are taking up too much of the eq. spectrum.  It could be a combination of both.  If you’re putting them at a level where they are perceived to be too loud compared to the rest of the instruments, it’s ultimately a question of proportion.  Eg.  If a vocal covers nearly as much of the eq spectrum as say, a drum kit, then the listener [I]perceives[/I] the vocal to be “larger” than the whole drum kit, which just doesn’t make sense & messes with our perceptions of “reality”.  Similarly, if a vocal has nearly as much dynamic range as a drum kit, then the drum kit will sound “small” by comparison.

Now, when I say “context dependent”, I’ll try to explain what I mean:  Eg – Take a song with very soft drums (eg. brushes) & smooth-sounding orchestral accompaniment… Because the drums are [U]not [/U]big & fat covering the whole gamut of the frequency spectrum from low to high (& thus seeming to create the perception of “width”), & are [U]not[/U] causing huge transients that create the perception of “height”, the vocals can be left pretty much “full range” & relatively uncompressed, because our brain perceives that they are in proportion to the rest of the instrumentation, due to it’s subdued & soft “background” nature.
This is an example of what I’m talking about:

On the other hand, in a pounding rock/pop/metal mix, the drums, guitars &  bass need to be perceived by the listener as BIG, so to put those same vocals up against that kind of backing would require that they be turned up so loud too be heard evenly, they would “dwarf” the instrumentation by comparison.  Likewise, their full range spectrum of frequencies would make the other instruments sound small by comparison.  So, the mixer in this case trims out masses of low end from the vocal, possibly some high end too, to make space for the cymbals, then compresses & automates it within an inch of it’s life so that it has virtually no dynamic range.  Solo’d, it would probably sound thin & weedy, but in the mix it can be turned up so that it well above the level of all the other instruments, yet it doesn’t sound out of proportion.

A good example is this mix:

From the Slate Cup, I know one thing:  The vocals are KING – you have to hear every word clearly, no matter how crazy & dense the rest of the mix is… In a dense rock mix, that means to me that reductive eq is the way to go – be ruthless & prune everything that isn’t necessary.  Eq wise – it needs to inhabit it’s own “space”, but in reality, that space diminishes in size the more dense the mix becomes.

Dynamics-wise, think about how “flat” a distorted electric guitar waveform looks.  Now think of how “loud” & “big” you want the guitar sound to be [I]perceived[/I]. With the vocal in this context, the only recourse is to compress, double compress, reverse compress & automate (or whatever) the vocal if necessary to get it to maintain it’s [I]proportion[/I] to the guitars.

Once you’ve got it to that point, & the proportion is correct – finding the right level is a fairly simple matter of adjusting the fader… & if in doubt use the “whisper-mix-level” test to check for vocal intelligibility.

Saved Comments


Jjsodarock – 09-01-2012, 07:18 PM
Report Post
Edit
Reply

>Very good examples. I am very guilty of trying to put to much of every instrument and vocal into my mixes. In fHumbles’ words, I do not use “subtractive Eq-ing” nearly enough for the type of mixes we have received in the S.D.Cup. 90% of the songs that we have recorded and mixed for the six years that we have been open, are acustic in nature, Guitars , piano, strings, vocals bothe lead type and choir. Not much rock type stuff, and virtually no drums. The S.D. Cup and you guys are helping me a great deal. Good article, Jojo

aj113′s Avatar
aj113 – 09-02-2012, 12:15 PM
Report Post
Edit
Reply

Very good, and heartening to see fHfH advising people to do what I pretty much already do myself. If we’re talking about rock/metal, the way to go is basically smash the living crap out of it, shelve it off as much as you dare without it actually sounding like someone singing down a telephone, then feed it back to the mix and have a listen. In terms of level, one general rule of thumb I use is to get the vocals to the same (perceived) level as the snare. i.e., listen to the snare, then the vocal and compare the two in terms of perceived loudness. Generally (and I know there will always be exceptions) you don’t want either one of these to be much louder than the other. That’s just something I discovered for myself, not sure if anyone else does it this way, but it works well for me. As it happens, the Paramore track in the article is a perfect example of this.

dudermn’s Avatar
dudermn – 09-03-2012, 11:29 PM
Report Post
Edit
Reply

Whisper mix level you say ??? For some strange reason… I feel like my clothes have been ripped out of my closet.

I find that compressed female vocals are wicked sexy. The higher the ratio the more porno do they sound. Though it could just be me.

Kinda thinking about it…..are there any songs out on the professional level that do not partake the use of compressed vocals ?

pkv2010′s Avatar
pkv2010 – 09-04-2012, 08:58 AM
Report Post
Edit
Reply

Quote Originally Posted by dudermn View Post
Whisper mix level you say ??? For some strange reason… I feel like my clothes have been ripped out of my closet.

I find that compressed female vocals are wicked sexy. The higher the ratio the more porno do they sound. Though it could just be me.

Kinda thinking about it…..are there any songs out on the professional level that do not partake the use of compressed vocals ?
In Australia we have a great mag called ‘AudioTechnology’. In the current issue (no. 89), Jack White (White Stripes etc) talks about his new album Blunderbuss, and he and his mixing engineer Vance Powell make a point of mentioning very little compression was used on the album.

pkv2010′s Avatar
pkv2010 – 09-04-2012, 09:00 AM
Report Post
Edit
Reply

Thanks ff, wonderful article!

shackman’s Avatar
shackman – 09-04-2012, 09:21 AM
Report Post
Edit
Reply

Excellent work FF.

I can’t say I agree with you regarding compression on vocals. I hear what you see, I see how you mean and I still don’t lie the idea.

I have always and probably WILL always take the long-winded route and automate rather than compess. You can arrive at exactly the same point when the vocal is balanced with the mix (or guiatrs or whatever) and then find the level you want simply.

My problem with compression is that at some points in a track I may want the vox to become almost whispered, yet audible and at others (i the same track) I may want screamers) With such scenarios, comprssion is MUCH more laborious to get right than a simple volume envelope.

I don’t think I’m right or wrong any more than I think you are. Horses for .. you know.

And anyways, I’m a artist/performer, not a mix engineer!

dudermn’s Avatar
dudermn – 09-04-2012, 12:29 PM
Report Post
Edit
Reply

Quote Originally Posted by pkv2010 View Post
In Australia we have a great mag called ‘AudioTechnology’. In the current issue (no. 89), Jack White (White Stripes etc) talks about his new album Blunderbuss, and he and his mixing engineer Vance Powell make a point of mentioning very little compression was used on the album.
Not a big fan of the stripes. Thanks for the heads up though !

I am wondering if Whitney Houston or Celine Dion run compressed vocals?

shackman’s Avatar
shackman – 09-04-2012, 01:28 PM
Report Post
Edit
Reply

Quote Originally Posted by dudermn View Post
I am wondering if Whitney Houston or Celine Dion run compressed vocals?
Doubt if Ms Houston is. She’s kinda …. DEAD!

dudermn’s Avatar
dudermn – 09-04-2012, 02:00 PM
Report Post
Edit
Reply

Quote Originally Posted by shackman View Post
Doubt of Ms Houston is. She’s kinda …. DEAD!
….How did I forget that ?
At least she finally gets to have Bonham play drums for her.

cporro’s Avatar
cporro – 09-10-2012, 11:55 AM
Report Post
Edit
Reply

interesting you talk about proportion. i usually say relationship. same thing i think.

i remember bob katz suggesting doing a mix with no compression. just automation. as an exercise. sounds like work! and andy wallace saying he uses compression for the sound of it..not really to control levels. interesting.

jack white…i always love the sounds he comes up with. but he’s doing something different then most of us. he’s recording with things that compress naturally like tape…and tubes, and stuff. right? so i’m not surprised he uses little compression.

garageband’s Avatar
garageband – 09-10-2012, 12:03 PM
Report Post
Edit
Reply

Quote Originally Posted by dudermn View Post
….How did I forget that ?
At least she finally gets to have Bonham play drums for her.
Cue the Righteous Brothers’ “Rock and Roll Heaven.”

pkv2010′s Avatar
pkv2010 – 09-10-2012, 11:35 PM
Report Post
Edit
Reply

Quote Originally Posted by cporro View Post

jack white…i always love the sounds he comes up with. but he’s doing something different then most of us. he’s recording with things that compress naturally like tape…and tubes, and stuff. right? so i’m not surprised he uses little compression.
yes, he is using tape and tubes.

Nanowire’s Avatar
Nanowire – 09-15-2012, 04:14 PM
Report Post
Edit
Reply

Quote Originally Posted by aj113 View Post
Very good, and heartening to see fHfH advising people to do what I pretty much already do myself. If we’re talking about rock/metal, the way to go is basically smash the living crap out of it, shelve it off as much as you dare without it actually sounding like someone singing down a telephone, then feed it back to the mix and have a listen. In terms of level, one general rule of thumb I use is to get the vocals to the same (perceived) level as the snare. i.e., listen to the snare, then the vocal and compare the two in terms of perceived loudness. Generally (and I know there will always be exceptions) you don’t want either one of these to be much louder than the other. That’s just something I discovered for myself, not sure if anyone else does it this way, but it works well for me. As it happens, the Paramore track in the article is a perfect example of this.
I too use the snare (and kick) as a gauge for the vocal levels on anything with a steady beat. It seems like a good starting point, but I never get it past sounding OK. Still learning on compression and (reductive) EQ. My neighbour is a sound engineer who does a lot of live sound where the LF/HF cutting is quite rigorous. I guess I read too many post where people preached caution with compression and EQ.

andresix’s Avatar
andresix – 09-16-2012, 09:12 AM
Report Post
Edit
Reply

for no compressed female vocals: try sinead o’connor, she’s known to HATE compression . Great article.

dudermn’s Avatar
dudermn – 09-16-2012, 09:50 PM
Report Post
Edit
Reply

will do !
Check out sinead o connor !

jrod9900′s Avatar
jrod9900 – 09-24-2012, 02:29 PM
Report Post
Edit
Reply

Great article. I can’t stress enough how important it is to hear and understand every word. I had a mix that I thought was one of my best, and I got hit with “I can’t understand that one word he said” kinda thing from a couple pros. I guess I knew all the words and didn’t think about it from that context.

Relative EQ and compression is king in this bidness. Pick your focal point, usually the lead vocal, and run with it.

Great post!

JROD

dudermn’s Avatar
dudermn – 09-24-2012, 03:51 PM
Report Post
Edit
Reply

and soon, everyone will have obnoxious vocals, so afterwards the instrument based composition of songs will be weeded out, and an audio engineer will only have to work on vocals for songs, and than only than, will the world discover ancient latin hymes.

The borderline between vocally driven and instrument driven is about 4db at 2k.

Ken J’s Avatar
Ken J – 09-27-2012, 06:18 PM
Report Post
Edit
Reply

Balance between vocals and the instruments is the key here. No rules but the song has to sound right to the listener in the end.

As a professional engineer, I track everything at a mean level no greater then -15dBFS. Sure we are going to have peaks from the kick that may hit -10 dBfs but the reasoning is to keep the levels down so the sum of the entire mix comes out around -4dBFS. For beginners, I suggest tracking dry. Remember that effects and compression can easily be added later. For someone who is experienced, sure you can add stuff on the way in but remember that once it hits the track, anything you did can not be undone unless you retrack. Never add anything to the master bus such as compression or limiting. Anything done should be done at the track level only. If levels get too high because of compression added at the track level, pull the faders back to a reasonable sound level. Never assume louder is better. When applying EQ, negative is always better then positive to set up the mix. Don’t be afraid to cut. Sure the overall mix may come out quieter but dynamic range may increase and there is no need to worry about hitting that dreaded 0dBFS point of ungodly digital distortion.

Also remember that monitoring at 85 dBSPL the fatigue factor of the ear is around 1/2 hour of critical listening where at 75dBSPL the fatigue factor is almost 2 1/2 hours. Mixing on headphones, fatigue can set in in as little as 1/4 hour or less. The household or workplace not to mention the whole neighborhood doesn’t need to hear what is going on in your mix area. Only you do so cranking up the monitors is never the answer. Ear fatigue causes misjudgements in decision making in the mix that is critical to the project.

There are no special rules to making up your mix. It just has to sound right to the listener. Balance so everything in the mix is clean and clear from the lowest bass frequencies to the highest tap of a cymbal in proper proportions. There is no mystery to making up a mix.

jpjeffery’s Avatar
jpjeffery – 09-30-2012, 04:18 PM
Report Post
Edit
Reply

I was listening to Pink Floyd’s Comfortably Numb t’other day and it seemed to me that the mix of the first guitar solo fits in with the overall message here. It’s not loud, it’s almost distant thanks to plenty of reverb, and yet, it’s as clear as a bell, fitting in to its own tonal space.

I’m still struggling to understand the terminology and what it means, but I think I understand the principle of establishing where an instrument fits in to the overall soundscape, and not just with volume and panning.

fakas’s Avatar
fakas – 10-30-2012, 08:02 AM
Report Post
Edit
Reply

I’ve found out that working the compression (of the master buss) on a range somehow between 1khz to 10khz can help the voice to “glue” better with instrumental part of the song, not only in terms of volume but as sounding much more entwined with instrumental. Also little compression triggered by the vocals and applied to instrumental can help the vocals/instrumental relation.

 

Brandon Drury

Posts Twitter Facebook Google+

Brandon Drury quit counting at 1,200 recorded songs in his busy home recording studio. He is the creator of RecordingReview.com and is the author of the Killer Home Recording series.
join

No Comments

Be the first to start the conversation.

Leave a Reply