Another thread about vocals

[Lunkhead]

Reist, this is in response to your message in the Nur Ein Round 2 thread. I figured we should carry on the discussion, but in a new thread.

A pop filter/wind screen is pretty useful, and you can get a really nice one for $40. This is the one I have:

http://www.musiciansfriend.com/product/ ... sku=421105

Ken recommended it to me, so you know it must be good.

"normalizing" is processing an audio recording so that the loudest part is adjusted to a certain volume (almost always 0db, which is what you'd want to use). If you keep your volume relatively consistent across your takes/tracks when your singing, and you normalize all your vocal tracks, your vocal tracks will probably blend together better and be easier to eq/effect/mix.

A "hi-pass" is just a type of EQ that lets high frequencies "pass" through, and cuts low frequencies. Again, I'm no expert, but I generally apply a hi-pass to things that are not bass instruments (things like vocals, guitars, etc.) to cut some of the low frequencies from them and help make them distinct from the bassy things (bass, kick drum, etc.). Use your ear to play around, but you could start off by trying applying a hi-pass at 150-200Hz. (The experts may want to weigh in here and clear things up if I'm providing bad/mis-information.)

Compressing the track is a whole big gray area, so play around a lot there. You can search the Web for some ideas of what settings to start with, eg. a ratio of 3:1-4:1, short attack time, auto release, etc. etc. The settings will depend a lot on your compressor plug-in. If your compressor software is hard to figure out, trying finding a plug-in with some good presets that you can use as starting points.

If you're doing doubled vocals, you probably want to keep them both centered, in my novice opinion. If they don't line up, you really just need to work harder on being able to nail down your timing when you sing, to be frank. Panning one track off to the side isn't really going to make them sound more in synch. I sometimes pan backup/harmony tracks away from the main track, but not a double track.

Of course, this is all just to get a basic, generic, "clean" vocal sound, and again these are not even "rules" for that, just suggestions/guidelines/things that seem to work for me. And, this will all get you only so far depending on your mic and preamp, and most of all your singing.

[Reist]

That stuff sounds pretty good. I'm not sure if I can do the high pass filter on the AW1600 but I'll try to figure that out. Does the pop filter clip onto any mic stand? Is it easy to clip on, or does it wear away at the stand?

[Lunkhead]

The Stedman screen I linked to above screws on to the mic stand. It's very easy to attach, position, and remove, but I suppose you could scratch your mic stand with it if you weren't careful.

[Adam!]

Reïst, lots of mics have a lowshelf pad of 6-10db, or sometimes even a highpass built in. Even if you're pretty sure yours doesn't have this option, check the manual: some have a button hidden inside that can be tricky to get to. Anyway, I just found a brochure for the AW1600 that contains a screenshot showing fully parametric EQ: this thing should be able to manage a highpass just fine.

I'd agree with Lunkhead on the merits of panning the doubling to the center, although some musicians (M.I.A., Bradsucks) use wide-panned vocal doubling as a sort of signature sound. When you double a vocal, one take will invariably be better than the other; this might be kind of obvious, but bump the volume on the good one and drop the volume on the worse one. That will increase the clarity of your vocals while still keeping the richness of the doubling, making you sound a little more pro at the same time.

It's weird, everywhere I look online I see people recommending light vocal compression settings. Things like "2:1 ratio, slow attack, ~3 db gain reduction"... well, I don't know what kind of music those settings are for (folk? the lightest of light jazz?), but if you plan to ROCK I'd say crank that sucker up. I find myself usually hovering around a 4:1 ratio, fast attack, auto release (like lunk said), but pulling down 12 or even as much as 20 db. Just obliterated. I don't know why, but in-your-face vocals just sounds exciting to me. If you feel like you are loosing your voice's natural dynamics, consider copying your vocal to a second track and blitzkrieging that with compression, leaving the first track uncompressed. When you mix them both together you will get some of the dynamics and energy of your uncompressed track as well as some of the solidity and power of the highly compressed track.

Do not be afraid of reverb or delay on the lead vocal. For a lot of people 'reverb' is synonymous with '80s', but it doesn't have to be. Use short (I like 0.4-0.6s) decay times for a subtly thickening, rock-y reverb. A little tap delay, timed to the tempo of the song, can also work wonders to thicken a vocal without drowning it. Both effects can be slathered pretty liberally on backing vocals.

[ken]

I'm not so sure about this 6 inches from the mic advice. I say about 3 inches is plenty. Really, get as close as you can without touching. Close = intimate. The nicer your room, the further you can get from the mic. If your room doesn't sound good, get up on that mic. Really though, you should test it out. Try a vocal a foot away, 6 inches, 3 inches, and up on the grill. See what each one sounds like and use the one you like the best.

With EQ I tend to high pass around 200 to eliminate mud and room noise, boost a bit at 1.5k for presence, and use a high self at about 10K for "Air". For back up vocals or doubles, I will high pass even higher up at 350, cut at 1.5, and no high shelf.

Compression is tricky and I honestly can't use most compressors. I am thankful that the UAD-1 LA2A compressor is awesome and easy to use. It's got two knobs. Crank them both straight up and everything sounds awesome. Can't quite hear the vocal? Turn the gain up a notch. Instant awesome. I hope your recorder has something similar to that. Find a compressor that works for you. I don't normalize anything. That is what compressors are for. Normalization is Blue's thing.

Maybe you could post some of the raw tracks from your song and we could check them out. Sometimes that is very telling.

Be well,
Ken

[Billy's Little Trip]

JB's trick is pretty cool and I've used it several times when I just couldn't quite get the ambiance or fatness I needed.
1. Record your vocals
2. copy that vocal track
3. move the copied track slightly forward until the desired delay between the two vocal tracks is achieved.

This works great to match up the delay of the other instrument tracks to get it sounding like the vocals are in the same arena.

As far as a high pass. Just roll off the lows at around 500 Hz down with your parametric EQ that your AW1600 already has built in. Roll off slopes will vary.
......or use an SM 58 and basically do nothing but sing into it and add the desired amount of reverb.

[king_arthur]

While I don't consider myself a great vocalist (in particular, any
time I try to "rock," I always get comments about how the vocals
just seem too wimpy for the song), but here are a few thoughts on the
actual process of singing:

- know what your vocal range is and put the song in the right key
for your voice. For example, I can sing up to a "D" comfortably.
So if I was going to record "Twinkle, Twinkle Little Star," the
highest note in that song is the first syllable of "little," which
is on the 6 note in that scale (do=1, re=2, etc.). In the key of
F, the 6 note is a D. So I would do TTLS in the key of F. I might
go down to E if I was doing it with guitar, since that's a better
guitar key. If the melody sat on that 6 note a lot in the song,
I might also move it down to E or even D.

- and the first thing I'd do on TTLS would be to lay down the drum
machine and one guitar, and then I'd record a vocal track just to
confirm that I've got the key right.

- I also do some tweaking of the melody (which, for me, is hardly
ever in its final state until I'm done recording it). If I know
that I want to add harmony vocals on the chorus, I may move the
melody down a bit so that I can sing above it without straining
too badly. I've had comments that my melodies almost always
start high and descend, so sometimes I'll deliberately do an
ascending line, or if I notice that I'm staying on one note a lot,
I'll look for ways to vary it up.

- when I'm recording vocals, I always type out the lyrics, in nice
big print, and pin them to the wall right in front of me (generally
a little bit above eye level, so that I'm not closing down my
throat to look down at the words). As I start recording, I add a
LOT of notes to that paper - lyric changes, what note in the scale
(1 - 7 ) each line starts on, etc. I note places where I am
tempted to breathe but shouldn't. When the paper gets too messy to
sing to, I update the lyrics file on the computer and print a new
version.

- part of what I'm working on a LOT at this point is the phrasing.
What word in each line should hit on the 1 beat? If you listen to
my "Empty Wednesday," I sing "There's just not enough" before the
first beat so that "goth" hits on the 1, and is thus positioned as
the most important word in the line. I actually noted on the lyric
sheet for that one that I had to come in on the "2" beat of the
previous measure with "There's" in order to get to "goth" at the
right time. (listen also to the end of the bridge on my "A
Variation on Two," where I think I come in SIX beats before the
start of the chorus, to sing "it's so much more than just a variation
on [two].")

- additionally, I make a lot of little lyric changes at this point
to make sure the singing flows with the rhythm... there's a line
in EW about "I was gonna [ask] her to go to the prom..." ([ask] hits
on the one beat) which was originally going to be "I was gonna ask
her if she'd like to go to the [prom]" but I just couldn't fit all
that stuff in before the beat, so I shortened the text and shifted
the accent. This is all very deliberate and when I don't spend a
lot of time working out this rhythmic stuff, I usually wish later
that I had. Totally apart from being on or off pitch, if the
rhythm you sing the words to helps the listener know what the
important words are, it matters!

- I think fluffy posted once that he pins up the lyrics in front
of him and sings them over and over until he can do them from
memory, and then does his final vocal takes that way. I keep the
posted lyrics in front of me the whole time, and use the written
notes to help me remember all the little tricks I worked out to
make the lyric work.

- I generally get my best vocals when I can do three or four
different sessions... the first round is just getting a decent
demo of how the song goes; the second time, I'm working on
how to fix whatever went wrong with the phrasing, fixing
lyrics that looked good on paper but don't work when sung,
etc. By the last take, I have a pretty good grasp of the
melody I want to sing.

- I use an Oktava condensor mic, which I position about forehead
level, pointed down toward my mouth, and I probably stand about
six inches from it. As with posting the lyrics high up, this
helps me keep my throat more open. I also tend to have a lot of
trouble with big wind sounds when I sing an "f" (sound, not note),
so having the mic high up like that gets it out of the path of
all the wind.

- as far as being able to just hit the notes... practice, practice,
practice. If you're willing to be seen in a church, try singing in
a choir for a while - it's a different vocal technique than rock,
but you'll learn some stuff about breath support and harmony that
will come in handy later. Having the song in the right key for your
voice helps, too - especially for rock singing, aim for the highest
part of your range where you can sing well.

Charles (KA)

[Hoblit]

JB's trick is pretty cool and I've used it several times when I just couldn't quite get the ambiance or fatness I needed.
1. Record your vocals
2. copy that vocal track
3. move the copied track slightly forward until the desired delay between the two vocal tracks is achieved.

This works great to match up the delay of the other instrument tracks to get it sounding like the vocals are in the same arena.

As far as a high pass. Just roll off the lows at around 500 Hz down with your parametric EQ that your AW1600 already has built in. Roll off slopes will vary.
......or use an SM 58 and basically do nothing but sing into it and add the desired amount of reverb.

There are delay effects that will do this for you and you can usually control it by units in ms. I have one that will take out the dry track on the left (or right) and only play the 'wet' (delayed) track... so its a fake stereo effect.

Personally, when I double vocals ...I ALWAYS separate them. I usually put them about 50 places apart. 25 & -25 ... right to left. If your doubling your vocals to make them strong or get a chorus effect (to cover up bad notes) it works perfectly... but it WON'T cover up timing issues. You have to pretty much be Johnny on the spot about that.

[jb]

JB's trick is pretty cool and I've used it several times when I just couldn't quite get the ambiance or fatness I needed.
1. Record your vocals
2. copy that vocal track
3. move the copied track slightly forward until the desired delay between the two vocal tracks is achieved.

This works great to match up the delay of the other instrument tracks to get it sounding like the vocals are in the same arena.

There are delay effects that will do this for you and you can usually control it by units in ms. I have one that will take out the dry track on the left (or right) and only play the 'wet' (delayed) track... so its a fake stereo effect.

Part of the coolness of "my" trick is that it doesn't take a plugin, so a) you don't have to figure out how to make the plugin do what you want and b) you don't have to have another effect taking up CPU.

Don't forget that when doing this, the final step is to pan the two tracks in opposite directions. You should play with the amount of panning-- for example, I will often do this trick with acoustic guitar and then pan the two tracks totally left and right. If I do it with vocals, I will often just pan them 50% L and 50% R. Taking advantage of, and filling, the stereo space. I think this trick would probably work fabulously for an organ patch too, if you want it really nice and fat.

Too much of this doubling, it should be noted, can be too much of a good thing. People, I've found, often like to hear just a single bare voice in a track. I think it gives the ears a bit of a rest from all the processed noise, but there's also just something about your voice, alone, with just enough treatment to clean it up and sound nice.

[Paco Del Stinko]

Although I don't enjoy hearing my own voice, I copy the lead vocal track and leave it in time with with the original, both panned hard to either side. Of course, it ends up sounding like it's straight up the middle, but it gives you plenty of level to work with, and it's easy if you link the tracks together. I typically add a touch of very short delay to slightly thicken, say around 3ms, and a bit of reverb. I like the JB method, and will employ that either in vocals or guitar at some point.

[Hoblit]

JB's trick is pretty cool and I've used it several times when I just couldn't quite get the ambiance or fatness I needed.
1. Record your vocals
2. copy that vocal track
3. move the copied track slightly forward until the desired delay between the two vocal tracks is achieved.

This works great to match up the delay of the other instrument tracks to get it sounding like the vocals are in the same arena.

There are delay effects that will do this for you and you can usually control it by units in ms. I have one that will take out the dry track on the left (or right) and only play the 'wet' (delayed) track... so its a fake stereo effect.

Part of the coolness of "my" trick is that it doesn't take a plugin, so a) you don't have to figure out how to make the plugin do what you want and b) you don't have to have another effect taking up CPU.

Don't forget that when doing this, the final step is to pan the two tracks in opposite directions. You should play with the amount of panning--

The luxory of my light weight (yet totally generic and unknown) recording/mixing software is that it comes with this effect and has it as a preset. But yeah, when I've used other programs to do the same thing it does get trickier.

As far as too much doubling & panning, you're right. Too much can be very distracting and if you separate just a hair too much...it'll give the listener a headache. (I've found myself getting dizzy when wearing headphones)

[Billy's Little Trip]

As said, JB's trick is cool because of the versatility of it. I mentioned it to Andrew because his rig doesn't support VST plug-ins, if I remember correctly.
As far as panning using JB's trick, I've panned out wide as mentioned, 25% left, 25% right, etc. But because of my mix down method, I've found that it gets too wide on the vocals and gets buggy on the ears. So I will either leave the L/R tracks centered, or widen them only 2% to 7% L/R, but always equal L/R on the panning.
As Paco said, just the fact that you copy the main vocal track, gives you a fair amount of level to work with even if you keep them both centered, so you can use one with delay and keep the other a bit dryer, then just move the wet track up under the dryer track until the ambiance matches up to the other tracks without loosing too much vocal clarity. But that's also because Paco has a very robust voice, lol. Us whiny guys need to fatten.

[jb]

A bit more on this trick:

When you've got the two tracks, you should pipe them both through a "group" track of some kind. In Cubase it's called just that-- a group track. This will let you manipulate both tracks as one, for EQ, for volume enveloping, etc. You can also then apply effects to both of them-- like compression, to make sure they both sound the same. You may also want to lock them together if possible. You will have two tracks to deal with, so it can be a little annoying when you want to adjust stuff.

[Lunkhead]

Yes, absolutely use group tracks if you've got them.

On the other hand, I personally prefer to keep it simple, and I never do any of the copy/paste/effect/etc. stuff mentioned above. If I want to make the vocals more prominent, I just try to make sure they cut through the mix, and if I want them "thicker" I just record another track. But those are just my preferences, as they fit the sounds I go for.

[Hoblit]

Yes, absolutely use group tracks if you've got them.

On the other hand, I personally prefer to keep it simple, and I never do any of the copy/paste/effect/etc. stuff mentioned above. If I want to make the vocals more prominent, I just try to make sure they cut through the mix, and if I want them "thicker" I just record another track. But those are just my preferences, as they fit the sounds I go for.

I don't have group track options on my software choice BUT I wouldn't use it anyways. I'm always taking double tracks and doing 'stuff' to them differently anyways. Sometimes one vocal track will be clean (with an eq roll off) while I'll use the other one for 'every other line' emphasis and give it the telephone/old radio trick, its so subtle that you can barely tell, but at the same time it enhances the song just a hair.

Using JB's trick on a guitar solo with a little further widening creates a great ping-pong delay effect that can make for a BLARING solo.

[obscurity]

but it WON'T cover up timing issues. You have to pretty much be Johnny on the spot about that.

Or use melodyne

[Reist]

Wow, guys. Thanks a lot - I'm blown away by the energy you're all putting into this.

I'm not sure if I've got the grouping options on the AW1600 - I would love using that for drums.

I stopped by Axe Music to pick up a pop guard (and new drumsticks - they're all blown), and a bit of advice for mics. This is what the guy said -

1. You don't want SM58s for studio recording.
2. B1s and all cheap Behringers suck.
3. Buy a CAD GXL2200. It's 89.98 with the shock mount, and looks cool.

Do you agree? Disagree? Just a reminder (again) - this is what the guy at Axe said. I'm just trying to make up my mind (for when I'm sure that I want a new vocal mic).

On my next songs I'll definitely try to work in some of those panning techniques, but I think it'd be good if I posted the raw tracks for a new song ... someone suggested that above. When I record my next one, I'll get on that.

Thanks again, guys.

[Steve Durand]

2. B1s and all cheap Behringers suck.

Don't confuse the Studio Projects B1 with the Behringer B1.

[jb]

I'm pretty sure I've seen a video of James Hetfield singing into an SM58 while recording Metallica's "Some Kind of Monster" album.

The right tool for the job is the right tool for the job.

I don't trust music store dudes at all one little bit.

That said, I love my CAD Equitek 100 microphone. I think that's what Jeff (of ADD) uses for everything. Dunno about the GXL thingy, but CAD is no slouch.

[Reist]

2. B1s and all cheap Behringers suck.

Don't confuse the Studio Projects B1 with the Behringer B1.

Oops. For some reason I had them fused as one in my brain. I've checked around a bit online, and they both seem to be quite favorably reviewed. Would you still recommend the Studio Projects one over the Behringer?

[Hoblit]

I'm pretty sure I've seen a video of James Hetfield singing into an SM58 while recording Metallica's "Some Kind of Monster" album.

The right tool for the job is the right tool for the job.

I switch between two microphones for vocals depending on the project. My SM58 is one of those mics and I personally swear by it.

[Caravan Ray]

I don't trust music store dudes at all one little bit.

No. Simply having a black T-shirt, body odour, long, greasy hair and 6 inches of visible arse-crack* does necessarily qualify someone to distribute technical advice.

(*I assume that "music store dudes" are identical the world over)