What's your vocal recording method?

richardyot

@dreamcartel said:

@richardyot said:
I usually record a scratch vocal and listen to it for a couple of days first, just to check I'm happy with the words and the melody. At this stage I will usually make some changes, there will be some lyrics that don't sit right and maybe also some tweaks to the melody after I hear it in context.

For the final takes my workflow is similar to yours: I sing each section separately and usually do around 8 to 12 takes that sit in a take folder in Logic.

The loop recording process in Logic makes this task really fast. When I used to do this in Auria I always had to stop the transport and return to the start of the loop and record again. Loop recording means you can stay in the zone and just work on the performance, without having to constantly fiddle with the transport controls.

I will then carefully listen through all the takes and choose the best one for the lead, and a second one as a double.

I don't always add harmonies, but if I think some are needed (usually in the chorus) I will sing them using the same workflow. I generally don't pan any vocals, except very occasionally the low harmonies.

I don't find the editing particularly tedious, in Logic it's pretty quick to loop a section and go through a take folder. I don't comp individual words (or even lines), I usually use the whole take (whole verse, whole chorus etc) without chopping it up into comped sections. So in terms of going through the takes the process is quick.

Later on I might need to do some editing of the audio regions to deal with sibilants, that process can be time-consuming. I will slice sibilants out into their own mini-regions and reduce the gain, and fade in and out to tame the problem sounds.

My room is very reverberant because it has wooden floors and wooden blinds, so I've built a homebrew vocal booth using acoustic foam stuck to some boards with bookstops glued to them as makeshift stands:

And I've been through a few microphones and interfaces over the years, but I'm currently settled on this, an SM7B and SSL 2. I find the SM7 very forgiving compared to condenser mics because it doesn't emphasize all the little pops and clicks from the mouth as much as more sensitive mics. In the past I would lose perfectly good vocal takes because there would be annoying mouth and throat noises in there, and the SM7 seems to mitigate that.

Wow, what did that run?

Also do you find using an sm57 with a pop filter adequate for vocals or should It be sm58

The SM7 was £330 and the SSL2 was £220 I think.

The acoustic foam was about £30, can’t remember what I paid for the board and the bookends, not much probably.

The SM58 is very similar to the SM7, I think it would work very well.

Ailerom

@Telefunky said:
A Pauly P120 pop filter may make a huge difference in front of a SM57.
At least in my case (it‘s depending on individual voice and/or air stream), I observed a clarity never heard before on that mic. Stunning.
(the P120 is very expensive, there a foam based Rycotes that are probably comparable for about half the price)

Imho the P120‘s „trick“ is 2 layers of randomly woven material. Under a good 20x magnifier there’s no pattern detectable.

That looks great. The one that came with my mic seem to do very little.

tahiche

I record with a condenser Mike, it’s a modded Oktava. I’ll do a couple of runs of improvised gibberish and pick out the bits I like. Then I’ll do some lyrics that fit the gibberish and record section by section.

I tend to record very close to the mic/pop shield. This has a nice proximity effect, but lately I’ve been noticing that it’s hard to place the vocal somewhere further back… Even with tons of reverb it sounds like an upfront vocal with reverb. I guess I have to move further back from the mic when I don’t want it upfront. I actually hadn’t thought about this until recently where no amount of eq or reverb would move the vocal back in the sound field… Any thoughts on this?. It makes sense that a vocal recorded close to the mic will have a lot of detail , but that same detail makes it inherently sound “close”, you don’t hear a singer breathing when he/she is further away and having the vocal so upfront is too predictable and a bit boring.

BroCoast

sm58>Space Echo.

I don’t do many takes, usually I get progressively worse. 😂

richardyot

@tahiche said:
I record with a condenser Mike, it’s a modded Oktava. I’ll do a couple of runs of improvised gibberish and pick out the bits I like. Then I’ll do some lyrics that fit the gibberish and record section by section.

I tend to record very close to the mic/pop shield. This has a nice proximity effect, but lately I’ve been noticing that it’s hard to place the vocal somewhere further back… Even with tons of reverb it sounds like an upfront vocal with reverb. I guess I have to move further back from the mic when I don’t want it upfront. I actually hadn’t thought about this until recently where no amount of eq or reverb would move the vocal back in the sound field… Any thoughts on this?. It makes sense that a vocal recorded close to the mic will have a lot of detail , but that same detail makes it inherently sound “close”, you don’t hear a singer breathing when he/she is further away and having the vocal so upfront is too predictable and a bit boring.

I have an Oktava, it's the stock MK319, it's a good mic.

Have you tried singing further back from the mic? It might solve the problem, most singers usually sing about 20-30cm away from the mic unless they're going for the proximity effect.

Ailerom

@tahiche said:
I record with a condenser Mike, it’s a modded Oktava. I’ll do a couple of runs of improvised gibberish and pick out the bits I like. Then I’ll do some lyrics that fit the gibberish and record section by section.

I tend to record very close to the mic/pop shield. This has a nice proximity effect, but lately I’ve been noticing that it’s hard to place the vocal somewhere further back… Even with tons of reverb it sounds like an upfront vocal with reverb. I guess I have to move further back from the mic when I don’t want it upfront. I actually hadn’t thought about this until recently where no amount of eq or reverb would move the vocal back in the sound field… Any thoughts on this?. It makes sense that a vocal recorded close to the mic will have a lot of detail , but that same detail makes it inherently sound “close”, you don’t hear a singer breathing when he/she is further away and having the vocal so upfront is too predictable and a bit boring.

That's a complicated one. I have my pop shield about 10cm from the mic. I record about 10cm from the pop shield. I do a few things to change depth. Sometimes I lean back, which helps if I'm screaming a bit more, if something is going to pop I try to direct it just a little to the left or right of the mic. If I'm recording something pop heavy I may step back to 30cm or more. For intimate I seem to have less pop issues because I am singing in a different way. I don't feel like I'm having any issue with positioning the vocal.

myapologies

Hi, does anyone have any tips if one was trying to record singing vocals into the standard internal iPad mic? What fx chain (order would be helpful) and other ideas might maximize quality? I have all the Fabfilters and more apps than necessary. Thank you!

abf

My wife sings. First we found the spot in the music room where it sounds the nicest when she sings, with the least amount of flutter echoes and such. It was easy as there was only one good sounding spot after all.

We record her by setting up two foldable room dividers at that spot. They each have four panels so we make a large octagon with them, leaving open a space so she can see me at the mixer, then we drape two puffy duvets over the top. The setup takes almost all of the room sound out of the mic. It takes about two minutes to put up, then the mic and the pop shield go on the stand.

When I record acoustic guitar I leave more space between the panels to get a little more room sound into the recording. The panels are nice looking and afterwards they just stack against the wall and take up no space.

We've used several mics over the years but settled on condensers because she sounds sort of congested recording through dynamic mics. My raggedy voice sounds better through a dynamic mic, but she can sing and has a sweet voice.

The main thing I do when mixing her vocal is volume automation. I use volume automation for the de-essing as well. That's the thing that makes her vocal track the best it can be- making sure every line is sitting just right. Compression is good but the compressor can't make the decisions I make about when to push the vocal up and when to drop it back.

Disclaimer- I'm no pro!

pedro

@richardyot said:
I usually record a scratch vocal and listen to it for a couple of days first, just to check I'm happy with the words and the melody. At this stage I will usually make some changes, there will be some lyrics that don't sit right and maybe also some tweaks to the melody after I hear it in context.

For the final takes my workflow is similar to yours: I sing each section separately and usually do around 8 to 12 takes that sit in a take folder in Logic.

The loop recording process in Logic makes this task really fast. When I used to do this in Auria I always had to stop the transport and return to the start of the loop and record again. Loop recording means you can stay in the zone and just work on the performance, without having to constantly fiddle with the transport controls.

I will then carefully listen through all the takes and choose the best one for the lead, and a second one as a double.

I don't always add harmonies, but if I think some are needed (usually in the chorus) I will sing them using the same workflow. I generally don't pan any vocals, except very occasionally the low harmonies.

I don't find the editing particularly tedious, in Logic it's pretty quick to loop a section and go through a take folder. I don't comp individual words (or even lines), I usually use the whole take (whole verse, whole chorus etc) without chopping it up into comped sections. So in terms of going through the takes the process is quick.

Later on I might need to do some editing of the audio regions to deal with sibilants, that process can be time-consuming. I will slice sibilants out into their own mini-regions and reduce the gain, and fade in and out to tame the problem sounds.

My room is very reverberant because it has wooden floors and wooden blinds, so I've built a homebrew vocal booth using acoustic foam stuck to some boards with bookstops glued to them as makeshift stands:

And I've been through a few microphones and interfaces over the years, but I'm currently settled on this, an SM7B and SSL 2. I find the SM7 very forgiving compared to condenser mics because it doesn't emphasize all the little pops and clicks from the mouth as much as more sensitive mics. In the past I would lose perfectly good vocal takes because there would be annoying mouth and throat noises in there, and the SM7 seems to mitigate that.

And here I am with a blanket over my head and dictafone set on lossless 😎

Edit: sorry didn’t realize this thread was from january

yellow_eyez

@pedro said:

@richardyot said:
I usually record a scratch vocal and listen to it for a couple of days first, just to check I'm happy with the words and the melody. At this stage I will usually make some changes, there will be some lyrics that don't sit right and maybe also some tweaks to the melody after I hear it in context.

For the final takes my workflow is similar to yours: I sing each section separately and usually do around 8 to 12 takes that sit in a take folder in Logic.

The loop recording process in Logic makes this task really fast. When I used to do this in Auria I always had to stop the transport and return to the start of the loop and record again. Loop recording means you can stay in the zone and just work on the performance, without having to constantly fiddle with the transport controls.

I will then carefully listen through all the takes and choose the best one for the lead, and a second one as a double.

I don't always add harmonies, but if I think some are needed (usually in the chorus) I will sing them using the same workflow. I generally don't pan any vocals, except very occasionally the low harmonies.

I don't find the editing particularly tedious, in Logic it's pretty quick to loop a section and go through a take folder. I don't comp individual words (or even lines), I usually use the whole take (whole verse, whole chorus etc) without chopping it up into comped sections. So in terms of going through the takes the process is quick.

Later on I might need to do some editing of the audio regions to deal with sibilants, that process can be time-consuming. I will slice sibilants out into their own mini-regions and reduce the gain, and fade in and out to tame the problem sounds.

My room is very reverberant because it has wooden floors and wooden blinds, so I've built a homebrew vocal booth using acoustic foam stuck to some boards with bookstops glued to them as makeshift stands:

And I've been through a few microphones and interfaces over the years, but I'm currently settled on this, an SM7B and SSL 2. I find the SM7 very forgiving compared to condenser mics because it doesn't emphasize all the little pops and clicks from the mouth as much as more sensitive mics. In the past I would lose perfectly good vocal takes because there would be annoying mouth and throat noises in there, and the SM7 seems to mitigate that.

And here I am with a blanket over my head and dictafone set on lossless 😎

Nice little soundproof “booth “ there man

From what I understand
The sm58 is basically the sm57 with a pop filter (?)

Ailerom

@myapologies said:
Hi, does anyone have any tips if one was trying to record singing vocals into the standard internal iPad mic? What fx chain (order would be helpful) and other ideas might maximize quality? I have all the Fabfilters and more apps than necessary. Thank you!

In the past I just record raw audio. If you are monitoring the vocal choose your own flavour but for me comp and reverb are enough for tracking. Post recording I find myself relying heavily on eq and comp. Depending on the environment, Brusfri can be a big help. I usually do spend most time with comp and eq and then add spatial effects to taste.

myapologies

@Ailerom, thanks for the helpful reply. Hadn’t thought of Brusfri. Haven’t used it in a long time

Phil999

my technique is pretty much like @Ailerom. The goal is to sing the entire piece in one go, but I might choose parts from different takes. Recent aqcuisition was a WA-47 jr and a UA Volt 476.

Fun fact: the last published voice recording of mine was done on an iPad Pro 2017 internal microphone. I was yelling, after 15-20 takes, and it still wasn’t quite good. But who cares? Somehow it worked alright. And besides, the sound booth was a tiny tent in the forest.

wim

For me anything works as long as there's a mute button involved and it's always on.