I am getting trying to get to grips with mixing predominantly urban genres.

There is one overing thing that is getting me, and that is EQing room for the vocal.

I.e. I like the sound of the instruments /drums in the mix but then I struggle with carving a hole for the vocal. I.e. I like the sound of the intro/instrumental but it all feels too dense when the vocal arrives and there is too much masking of the vocal.

Are there any guidelines for cutting certain frequencies for male and female lead vocals in the backing? I realise muting elements helps, but if I want to keep the arrangement as is where should I be looking?

Are there any specific areas of the frequency spectrum that are more important for intelligibility that other and therefore starting in those areas when applying cuts to pads/strings/synth bass etc. to make more room for a vocal.

I have been thinking most vocals lines never go over an octave and a half in range so maybe I should make take this into account with the width of any cuts I am applying.

From listening to a range of male singers it seems that the most important area where I get the main essence of the vocal is centred around 1100Hz ish with a bandwidth of just over an octave required so this is one of the areas I have been experimenting with focusing my cuts on background instruments.

Any advice on the topic would be appreciated.
K

If you're struggling to fit the vocal in over the top of everything, have you tried starting first with the vocal and bringing everything else up around/underneath it? You might find this easier, as the vocal will always be the main part you're shaping things around

--------------------
www.myspace.com/michaeldow www.myspace.com/portasoundband

Yes I have tried that, but working that way I still have to cut out a hole for the vocal just the same. Might be better psychologically though as opposed getting attached to the sound of the backing then dropping the vocal.
G

I haven't mixed a lot of rnb stuff, but I have done a bit.

on the stuff I've worked on, the producer has given me a fairly simple backing/guide track, I would record the vocals to the guide track. The producer would take away the comped vocals and re-arrange and write the music to fit around the vocals.. and then bring all the files back to me for mixing..

basically the music is arranged around the vocals.. and when it comes to the mix I never have any difficulty making the vocals fit.. I never think about carving out space, as the space is already in the arrangement.

--------------------
www.Lozjackson.com

It doesn't have to be RnB specific I would have thought the principle of carving out a vocal is similar in rock or trance etc. i.e. a male or female vocal would still occupy a similar register and any masking of the vocal would need to be dealt with in a similar way.

To pick an example the chorus of Rhiannas umbrella or rude boy or loads of her tracks really all contain some pretty heavy scoops.

Just wondering if there are any clear areas that are more important than others i.e. to me on male vocal the 1k region seems more important not be be masked by synths etc. in order to make the vocal clear etc...
G

Quote foge:

---

I am getting trying to get to grips with mixing predominantly urban genres.

There is one overing thing that is getting me, and that is EQing room for the vocal.

I.e. I like the sound of the instruments /drums in the mix but then I struggle with carving a hole for the vocal. I.e. I like the sound of the intro/instrumental but it all feels too dense when the vocal arrives and there is too much masking of the vocal.

---

Don't carve it out, leave room for it in the arrangement in the first place. If some other instrument is competing with the voice for attention, leave it out of that section. Basically, write better music, don't polish turds!

I would agree leaving room in the arrangement is part of it,

however I am sure you will agree its normal practice to use subtractive eq to solve conflicts between say a kick part and a bass guitar but you wouldn't suggest the bass player doesn't play or that there composition is a turd.

Its also quite normal to have to carve out string/pad parts to make more room for a vocal just wondering what other people thought. Doesn't make the music a turd. I am trying to get a firm handle on which areas of a vocal part are most crucial open a mix around it.
K

Yeah but eq'ing the kick and bass is a lot easier if the arrangment and performance of the parts are well suited to each other. IE a driving solid bassline suits a driving kick drum whereas a sparser funkier part allows for more variation in the bass part.

As to equing you will find you get some definition in the 1-3k range and the body around 200-500hz try high pass filtering some stuff and making small cuts around the 2k range and see if that helps.


--------------------
Squarehead Jam Jar Facebook Jam Jar

Edited by JamesSimpson (12/09/10 01:46 PM)

I'm no urban expert, but the principles remain the same...

It's *primarily* an instrumentation/arrangement issue. With kick & bass, there are only two potentially competing elements... and you'll probably high-pass or low shelf anything else like guits or piano that fights with them. Same goes here... the main thing is not to mask the vox too much. E.g. writing a part that pre-empts or reacts to the vocal can be very effective, whether with guits, strings, synths, bvox or whatever. Just don't have too many things going on at the same time (same section yes, same time, no.).

When it comes to balancing, you can eq vox for more bite at 2.5kHz (ish) and more body 500-750Hz (ish, again). Or you can EQ those frequencies out of other stuff. You can also try automation or side-chain compression to get any offending/clashing parts to duck by a couple of db every time the vocal sounds. That way, you don't lose their overall impact, but there's a subtle shift towards the vox when they come in.

Don't forget the verbs, delays and low-pass EQs, all of which can be used to effectively push something back in the mix, which, by virtue of contrast, can help another element (lvox) to take centre stage. So say there's a big synth part playing in the chorus, you could automate a low pass EQ just to dull it down a bit while the lvox need to come to the fore, and bring the top end back in when they don't

EQing FX returns can be a help too, if cutting frequencies on the main instrument makes it not work — you can still keep that instrument, but reduce the frequencies in any given band bouncing around in a wash of reverb or delays.

Failing all that, a harmonic exciter on the vox *might* help it to cut through, but approach that with caution.

As always, there are many ways to approach this particular problem, what works on each track will vary. One trick may do the job, or it may be a blend of all the above.

Quote Exalted Wombat:

---

Don't carve it out, leave room for it in the arrangement in the first place.

---

Wombat is making an excellent point. If your arrangement is covering the vocal to the point where you need to do any serious 'carving' then it needs some attention before you start hitting the EQ.

When I'm building a mix I'm cautious with EQ boosts until I know I have room for everything in the arrangement. I'm initially more interested in EQ cuts, particularly filters. It's at this stage also that I become ruthless about musical parts - as long as I can convince the artist!

For me it's in the 100-200Hz range where the battle for the vocal often begins. It's back to filters if there's anything treading in this territory that needn't be there. At this time I'd rather the guitarist wasn't around until I've finished working my magic! If I can get this part of the mix right then it usually begins to fall into place.

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

Quote foge:

---

I would agree leaving room in the arrangement is part of it,

however I am sure you will agree its normal practice to use subtractive eq to solve conflicts between say a kick part and a bass guitar but you wouldn't suggest the bass player doesn't play or that there composition is a turd.

---

Well, I'd LIKE to be able to say "No, this would all have been sorted out by the players - all you had to do was put a microphone in front of it!". But I have to be realistic and accept that a lot of music these days is constructed from pre-recorded loops and "beats" that were not custom-made for that song, and that there's a concept in some styles that once a two-bar loop is established, it HAS to continue ad nauseam to the end of the song :-) (Yes, I exaggerate, but not THAT much!) So a certain amount of turd-trimming may be necessary to make it all fit together.

In my (limited) experience it is usually best to arrange the tune with the vocal part in mind. Then the parts that would clash with the vocal simply are not there at the wrong place; this means you can process the vocal just to make it sound good rather than having to do the corrective mixing.

There are a few more options beyond carving a valley for the vox with the EQ....

You could try spliting the vocal feed and having a heavily compressed second vocal behind the original. This, parallel compression, can make the vocal more solid and audible whilst retaining as much of the dynamic as you need.

You could put a compressor across a stereo sub mix of the backing, or part of the backing. Then send a feed from the vocal to the side chain input of the compressor. This way the vocal will trigger a reduction in the volume of the backing, or the part of the backing that is clashing with the vox, but only when the vocal is sounding.

With something like this it might need several subtle little tweaks rather than one big tweak...

You're going philosophical here, youre gonna have as many different opinions as different responses..

Its definitely something you need to consider pre-production, when arranging, dont overload the mid-range, keep the band / orchestra balanced.

But definitely, yeah, when I start mixing (and i never start with leads, i always start with drums + Bass), I always carve out room in advance for my lead in the mid-range areas, as low as 300Hz, or even lower, depending on the lead, especialy if its a male vocal, and as high as 1-1.5 KHz. it really depends on the materials.

I try to define backking harmony and rhythmic layers higher than the crowded low mid areas, and, well, the bass layers too, though obviously i give them presence in the low range.

even supporting counter-melodies should not fight over Spectral range with your lead singer, this way they can get the level they deserve and it wont crowd your mix.

that's my two-cents..
Boy

--------------------
The Man in the Crowd with the Multi Colored Mirrors on his HobNail Boots

Quote thefruitfarmer:

---

You could try spliting the vocal feed and having a heavily compressed second vocal behind the original. This, parallel compression, can make the vocal more solid and audible whilst retaining as much of the dynamic as you need.

---

Thats a very good input here !

--------------------
The Man in the Crowd with the Multi Colored Mirrors on his HobNail Boots

@ boywander

Parallel Compression is something I have found makes a big difference when mixing vocals.

Quote thefruitfarmer:

---

@ boywander

Parallel Compression is something I have found makes a big difference when mixing vocals.

---

Yeah, me too... though I tend just to use a comp with built in wet/dry control rather than bother multing out to a separate track. DDMF NY Compressor is pretty good for this, as are Jeroen Breebaart's Red Phatt Pro and of Softube's FET Compressor.

I often get songwriters who have thought up a million little parts that they have though of, shall I say serially, but expect to hear each and every nuance in parallel in the mix. Funny thing is, we hear many records and remember little but the vocal.

If you have mixed the perfect instrumental, that's a good reason why the vocal doesn't sit. Listen to most well-mixed tracks and individual instruments sound kind of thin but the overall impression is good. If you mix each to sound perfect in isolation they sound muddy and aweful together. (But don't ask me how Tuesday Night Music Club was mixed - that's a mystery to me.)

--------------------
Daniel Davis
Edinburgh Recording Studio Windmill Sound

Quote Daniel Davis:

---

I often get songwriters who have thought up a million little parts that they have though of, shall I say serially, but expect to hear each and every nuance in parallel in the mix. Funny thing is, we hear many records and remember little but the vocal.

If you have mixed the perfect instrumental, that's a good reason why the vocal doesn't sit. Listen to most well-mixed tracks and individual instruments sound kind of thin but the overall impression is good. If you mix each to sound perfect in isolation they sound muddy and aweful together. (But don't ask me how Tuesday Night Music Club was mixed - that's a mystery to me.)

---

I just dialled up "Tuesday Night.." Very nicely done, but it seems to me the mixing was done largely in the arrangement (just the way it ought to be!) Can you point me to a track where you feel mixing technique solved a problem particularly well?

Quote Mixedup:

---

though I tend just to use a comp with built in wet/dry control rather than bother multing out to a separate track.

---

Well, though I'm quite new to parallel compression myself, still, IF you want two different degrees of compression running in parallel, you can't use a wet/dry controller. you can serially chain two compressors, and use the wet/dry on the latter, but it won't be the same.
I run an educational facility here, and many of the singers are inexperienced, and as much as I want to retain some dynamic range, I cannot use a completely dry uncompressed feed..

--------------------
The Man in the Crowd with the Multi Colored Mirrors on his HobNail Boots

Edited by Boywander (15/09/10 10:28 AM)

Quote Exalted Wombat:

---

Can you point me to a track where you feel mixing technique solved a problem particularly well?

---

I think in any mix of electronic oriented materials, house, dance, what have you, where many of the sources are lush rich synth sounds, any sound man has no choice but to use a variety of mixing techniques (corrective EQ included) to have the lead stand out.
Try soft examples like Daft Punk, Lamb or Air, serious mixing work here, even after terrific arrangements.

--------------------
The Man in the Crowd with the Multi Colored Mirrors on his HobNail Boots

Quote Boywander:

---

I run an educational facility here, and many of the singers are inexperienced, and as much as I want to retain some dynamic range, I cannot use a completely dry uncompressed feed..

---

If you're using compression to fix poor mic technique, don't you owe it to your students to TEACH them better mic technique?

That leaves you free to compress for a more punchy sound - probably inevitable in modern commercial styles of music.

Mixing should be about enhancing a good performance. Too much of the discussion here is about rescuing a bad one. Fix it BEFORE the mix!

Quote Exalted Wombat:

---

If you're using compression to fix poor mic technique, don't you owe it to your students to TEACH them better mic technique?

---

you are perfectly right my friend! proper mic technique is very important just as properly controlling the singing voice, and proper diction, hell yeah!
But, you know, young singers do have a learning curve.. and even the most experienced singer, many times, when she whispers for example, will need compression not only for sound shaping, but simply for thinner controlled dynamic range.

--------------------
The Man in the Crowd with the Multi Colored Mirrors on his HobNail Boots

Quote Boywander:

---

Quote Mixedup:

---

though I tend just to use a comp with built in wet/dry control rather than bother multing out to a separate track.

---

Well, though I'm quite new to parallel compression myself, still, IF you want two different degrees of compression running in parallel, you can't use a wet/dry controller. you can serially chain two compressors, and use the wet/dry on the latter, but it won't be the same.
I run an educational facility here, and many of the singers are inexperienced, and as much as I want to retain some dynamic range, I cannot use a completely dry uncompressed feed..

---

So you're doing Brauer-esque compression? ie. multing out to several tracks and using a different compressor on each? Nothing wrong with that if it works for you: I find that useful where I want a different effect in different sections of the song, but less so for general dynamics control on individual tracks — unless I'm wanting to EQ the parallel compressor or something.

A decent comp with a wet/dry control and a high-pass filter, a limiter, and if needs be, a multi-band compressor so you can target compression in certain areas should get you near enough. Or use two comps in series, one nailing down stray peaks, and the other just gently gluing things together. Dynamic EQ can be great too, for problem areas.

...and yes, get it right at source and all that: great if you're producing the whole thing, but a dedicated mix engineer won't always have control over that.

Quote Exalted Wombat:

---

If you're using compression to fix poor mic technique, don't you owe it to your students to TEACH them better mic technique?

---

It takes a while to teach mic technique.

When I have had inexperienced singers over I have tried various tricks to keep them in the sweet spot including putting the pop shield against their nose, putting their lyric sheets in front of them so they stay in the sweet spot.

I have taught a few people where their sweet spot is but what I have found is that they need to already know this before making a decent performance. I have had better results when someone does a spirited perfomance with the SM 58 on their lips and giving some compression on the way in rather than fannying about with the sweet spot on a LDC mic.

IMO and IME it is often better to somehow capture the performance rather than turn the recording session in to a teaching session. That said, if their technique is really that bad then you might as well make it a teaching session. Otherwise you will have your work cut out processing an "on axis / off axis" vocal......