I need help with live vocal monitoring, please be my christmas miracle

[edd]

I cannot sing using my headphones only with my tlm 103 running through my apollo twin x/ 1084 in unison, it's either too quiet and I cannot hear myself or it will clip when I belt.
For example :

He does a FULL take, with both quiet AND loud parts, how? Keep in mind I'm not talking about mixing/ mastering, just live performance.
I know this is a silly example, but when using airpods pros with the transparency mode, it just "works", you just hear yourself how you'd hear yourself with no headphones on(although I'd prefer a bit louder with instruments on). I want the same effect (or better) and I'd appreciate your help, please!

PS: I think they really didn't use much compression, was it because of the microphone used? Are the OG SM7's good at "blending dynamic ranges"?

 

[flandybob]

Not sure what you are trying to do but MJ had great vocal and mic technique. I’m also pretty sure they would use compression while recording, and then some more during mixing

 

[Joe Porto]

What headphones are you using with the Twin?

 

[bigshinyT]

Was about to ask the same re headphones plus, are you using a pop sheild, mic technique, compression? What's your gain staging look like? Maybe some screen shots or a vid of you recording your vox.

 

[edd]

I'm using Beyerdynamic dt 1990 pro headphones, the issue is making my voice be loud enough compared to the instruments for both quiet and louder parts. The issue is that high gain will clip and using a compressor will overdrive it when I belt and it just doesn't sound natural. Or I could just lower the instruments track to a very low level then crank the interface's volume knob, but I really doubt that's how it should be done... I don't know how much the 1084 should be driven? Because I tried being at around -16 db but then I can't hear myself/
Basically the dynamic range while monitoring is waaaay to big.
I didn't attach screenshots because I want to know the "right way" to set it up for that kind of vocals, because I've tried everything on my own, doesn't work good enough. I don't have a pop shield though.

I'm a novice so apologies if I'm struggling to explain well enough!

 

[dimestoremedicine]

Try one earphone off. Pan important stuff to the ear that's still on. Works wonders for most!

 

[flandybob]

I don't see too many options, off the top of my head:
- you have someone engineering for you and he/she "rides" the gain depending on the section of the song (whether singing quiet or belting)
- you sing in sections and adjust the gain depending on whether you are singing quiet or loud
- you use a compressor (you don't need to print it, it can be only for monitoring, but if it's working too hard it can throw off your performance)
- you "work the microphone", singing closer for quieter parts and taking a step back for louder ones (that will affect the tone though)

That's actually quite a bit of options

Then there's vocal technique, singing quieter or belting doesn't necessarily mean the actual volume of your voice is that huge of a difference. I personally don't know how to do that, I'm quite a poor singer, but accomplished one can manage this in some way, in conjunction with good mic technique

Or I could just lower the instruments track to a very low level then crank the interface's volume knob, but I really doubt that's how it should be done...

I don't see why not, I do this all the time. The volume pot is there to be used, I lower the volume of the track all the time and keep the vocal fader at 0 in console, then adjust the overall headphone volume to taste.

 

[Joe Porto]

The DT990 are not great cans for tracking. They excel at mixing but they are open back (mic bleed) and relatively high impedance (250ohms) so aren’t going to be super loud.

The closed back 770s in 32ohm would be a much better choice for tracking. Or even a cheap pair of Sennheiser 280pro closed back at 64 ohms will give you better volume and isolation.

On top of that, monitoring through a compressor to reduce dynamics should get you what you want.

 

[bigshinyT]

What @Joe Porto said re the higher impedance 990’s as def not ideal for tracking as require a good headphone amp for volume.

Do a test run with just your vox singing the loudest part and, using compression and checking your gain staging, get it to peak no higher than -6db. Now do the same but this time bringing up the backing tracks in the mix until you have balance between your vox and instruments that works for you. My guess though is that you’ll benefit greatly from swapping out your current headphones for lower impedance ones as Joe pointed out.

 

[Eric Dahlberg]

Try one earphone off. Pan important stuff to the ear that's still on. Works wonders for most!

Highly recommended. This was probably the first ever "studio trick".

 

[edd]

I tried with some IEMS and it still doesn't seem right, I still have the same issue, the dynamic range of my voice is too wide compared to the instruments that stay at the same level. For example here's another clip: https://www.youtube.com/shorts/0x0-fnGX5mQ
He's singing both soft and belting, screaming, his mouth glued to the microphone with headphones on. How do you think they achieved that?

 

[FutureLegends]

I tried with some IEMS and it still doesn't seem right, I still have the same issue, the dynamic range of my voice is too wide compared to the instruments that stay at the same level. For example here's another clip: https://www.youtube.com/shorts/0x0-fnGX5mQ
He's singing both soft and belting, screaming, his mouth glued to the microphone with headphones on. How do you think they achieved that?

He's one of the best singers ever and he's recording in one of the best studios in the world and he's having one of the best audio engineers in studio history sitting in the control room. Try that and see if it works better for you then!

I think he's singing louder than you think during the softer parts and quieter than you think when he's belting. You probably need to improve your singing and microphone technique and use compressors on your voice... I tend to see the microphone as an instrument, you need to some degree learn to sing so it sounds good through it. Sometimes that can be different from how you're used to sing...

 

[alwaysbecreating]

First disclaimer: I am NOT an expert - so what I'm about to say is just what has worked for me. This is also restating some of what has been said before - just with few more words.

Also - listened to the first vid, watched the second. It does sound to me in the "Thriller" vox like MJ is possibly moving back from the mic a little during the louder parts. Just something about the sound seems a little more "distant" to me in those moments. That all being said...here's how I would break down your problem:

1. The level of the mix (you want this where you want it, for comfort)
2. The level of your RECORDED vox (you want this to not be clipping, and to have whatever quality you want the unison preamp to be imparting)
3. The level of your vox in the mix, so that you can hear yourself no matter how you're singing.

So - let's attack those one at a time:

1. Level of the mix. This one is easy. Set it where you want it.

2. Level of the RECORDED vox. This is more subjective. Generally I think you would want to set this based on your loudest vocals so that it has the color (and doesn't have the clipping) that you want. This means that there will be a lot of dynamic range in what you record (unless you print compression). That's because your softest notes will not drive the preamp the way your loudest notes will. But they will still be recorded - and you will adjust dynamics for the soft parts as you mix (later in the process).

If you want your softest notes to drive the pre-amp, then you'll need to record them separately (so that you can turn the gain up). This is would be totally normal to do.

OR (not sure if this is correct) you could apply compression on the way in, to level your signal across the dynamic range. Or maybe you could use something like the Manley Voxbox (which has compression integrated) instead of the Neve 1084 (which doesn't). There was a good UA video on using the Manley Voxbox to record hip hop vox which showed a bit about how to do this.

You definitely want a pop filter so you can get close to that mic without causing plosive problems.

3. Level of your vox in the mix. You have some options here. In terms of the dynamic range of your voice, I would probably opt to apply compression on your vocal to what you're monitoring (i.e. not to what you're recording - you can choose). You probably already know this, so forgive the oversimplification, but compression will allow you to boost your overall signal (i.e. bring up the volume of the soft passages), but will keep your loud passages from blasting you (because they're being compressed). So the effect in your headphone/monitor mix is that your soft passages will still be loud relative to the mix, and your loud passages will be compressed. Whatever effects compressing the loud vocals creates, as long as it's ok for you to hear, won't be printed to the recording. You do also have a fader where you are controlling the level of your send to the monitor mix (without affecting the input gain that you're recording). After you get the input gain where you want it (step 2), then you just need to do some experimenting with the compressor and the fader that sends to the monitors, and you'll probably find something that feels good in your ears with some experimentation. Plus having a pop filter will let you work the mic, especially if you lean in for the soft passages.

Hope that's helpful! That's what's working for me so far. I'm learning as I go.

 

[slamthecrank]

A couple of things (some things others have already mentioned) --

1) His mic technique in the Thriller example is great b/c he's backing pretty far away from the mic in the super loud parts; then getting closer on the softer parts.

2) There is A LOT of compression going on. Fast compression, at that. You can even hear the compressor "let go" in between phrases when the mic picks up the bleed from his headphones.

3) I would never advise comparing anything I was doing to what Michael Jackson, Quincy Jones, and Bruce Swedien were doing during this period. They were at the top of their games, individually... and collectively putting out some of their best work... and using multi-multi-million dollar facilities to do it.

4) As far as the Thriller vocals go (I didn't listen to the other link you posted) ... there are definite "takes" involved. You can pretty easily hear the cuts and cross-fades. In other words, he didn't do that lead performance in one take or possibly even in one sitting. Don't expect yourself to make a recording like that at once, either. Give yourself some patience.

(and here's some levity .. possibly blasphemy to the Jackson fans ... if you want to hear an even stronger mic technique, check out what Prince was doing during this time. And then realize that he was pretty much playing all of the instruments as well. But he also surrounded himself with extremely competent engineers etc. It sometimes takes a village, so give yourself some breathing room).

 

[edd]

Hey guys, I'm sorry I wasn't trying to be mean or arrogant I'm just trying to explain my issue. I want to get better at singing with a microphone and headphones before going to the studio because the type of music I sing requires mic for projection, but for that I need to be able to do it at home, I'm not a mixing engineer, I did not compare myself to anyone, I was merely using the best example that I know of so that everyone could understand me.
Keep in mind I'm not getting much louder when belting it's just projected much more, but it clips if the Neve is set anything higher than -30db. That's fine really but then I can't hear myself singing the "quieter" parts. I was asking about the compression because "I didn't think" they were using lots of it and remember having read that somewhere, but like I said I'm a novice, I'm not an engineer, just trying to get a working vocal chain at home.

So I'm going to try setting the 1176 to 1 release, 2 attack and 4 ratio. How much should I aim for on the VU-meter for the highest parts? How about the soft/mid parts? Is it normal if it goes all the way up as soon as the belty notes "begins" then it starts going down?

 

[FutureLegends]

What is the typ of music?

If you set the compressor so it doesn't record the effect, it's not so important how you set the 1176... do slow attack and fast release and I wouldn't worry if it compresses between -10 and -20 at the peaks... as long as you get to a comfortable level.
Sometimes adding a bit of reverb can help as well...

Check the meters on the interface so you're not clipping already there... if you want to be able to turn the 1084 up more for the sonic flavour that adds, turn the -20 pad on...

 

[hotspot]

This is exclusively a gain staging and monitoring problem.
Curing this with a compressor is not the right approach.
At least not if you want to learn and improve.
Get to the root of the problem.
If you need help with this, we’d need at least more information, I.e. screenshots of your settings.

In the simplest case, just turn down the playback on the master fader. Problem solved.

 

[Eric Dahlberg]

Keep in mind I'm not getting much louder when belting it's just projected much more, but it clips if the Neve is set anything higher than -30db. That's fine really but then I can't hear myself singing the "quieter" parts. I was asking about the compression because "I didn't think" they were using lots of it and remember having read that somewhere, but like I said I'm a novice, I'm not an engineer, just trying to get a working vocal chain at home.

dimestoremedicine already provided the answer for hearing yourself during the quiet parts. Take off one ear of your headphones.

And yes, the difference in dynamic response between your TLM103 and the SM7 is MASSIVELY different. You will need to exercise even greater mic technique (facing off axis for the loudest parts) than MJ did. Maybe consider getting a dynamic or ribbon mic if you are not able to achieve good results with the TLM103. The Universal Audio SD-1 would be an excellent choice since it models other mics.