Recording and mixing vocals

ewal

I'm rubbish at recording and mixing vocals. And I want to get better at it.

This could well be because I'm rubbish at singing, but I don't think it's just that. Stylistically I'm typically making fairly dense guitar/synth based music with not a lot of space in the mix - liberal amounts of reverb/delay on guitar and keyboard parts etc. The overall effect is convincing enough without too much effort eq'ing and compressing instrument tracks. Until I add a vocal.... I struggle to make it fit. I know eq and compression are important. I know that I can add colour using different flavours of distortion. I know I can use reverb and delay. I know slapback delay can be useful and double-tracking might be an option too. But as yet I don't have a workflow or effects chain I'm comfortable with.

I really don't want to go on equipment buying journeys. I've got a decent condenser mic (Rode M3 I think), a capable audio interface (Audient EVO 8), and a good collection of effects to work with (Soundtoys, FabFilter, Helix Native, Valhalla Vintage...)

I guess it's just a case of keeping at it until I learn what works for my music and my voice? Maybe with a particular reference track in mind?

Anyone got any wisdom to share? Was there a point where it clicked for you? Was it really getting to grips with eq and compression techniques?

theatreanchor

YouTube mix with the masters. 

thecolourbox

I'm not great at it, but I think i've got better just by doing it more often, and then saving my track presets so next time I can think...I kind of want a similar sound to what i did on XYZ track and boom, loads straight up. You can get forensic with EQ frequencies or you can just play around with any old crap and eventually something sticks

ewal

theatreanchor said:

YouTube mix with the masters. 

I find them a little too advanced for me. Which probably says more about me than the channel.

Currently wondering if a different microphone might be a little more forgiving in my non-treated recording room.

AntonHunter

Different problem (specifically for live recordings with lots of spill), but I liked the idea behind this approach. Basically if you're struggling to get the vocals to sit into a busy mix, start with the vocals and fit everything else around them?

AntonHunter

Do you have a mix we can listen to for specific ideas though?

Danny1969

Room and mic wise you can get a reasonable vocal by building a makeshift vocal corner. Just stack sofa cushions in a corner and sing into the corner close to the cushions. I had a large print picture I also use which is stuffed with rockwool. You can buy these cheap then and then just go round the wood frame again to make it deep enough to take 25mm or 50mm rockwool

Compression, EQ and automation are the 3 main things to get a vocal to sit in the mix. Here's a very simple guide as a starting point. If you let the missus pick the print then it can stay on the wall as covert vocal treatment.

Put on an EQ plugin. Set the hi pass filter to 100Hz .... make a cut around 300 to 500Hz with quite a low Q, so a wide gentle scoop out then make a slight boost with a low Q around 6K

Put a compressor plugin on and set the ratio to 5:1 with a medium attack. Then lower the threshold until you see around 6dB of gain reduction on the loudest bits of singing.

 Set vocal channel automation to write and manually ride the fader once through. Then set it from write to touch and go through again correcting any errors and improving .

That's generally enough to get any vocal sitting volume wise roughly right in the mix.

For a effects I advocate making aux buses and feeding the vocal to them. I normally have a reverb aux send and a delay aux send. The delay is timed to the BPM. Then I write automation on the aux send fader to the reverb and delay

But everything else also needs to be treated with the same attention because the vocal will never sit right if something else like a bass guitar or snare drum isn't totally consistent in volume when it needs to be. Again,  compression and automation are the key.

ewal

AntonHunter said:

Do you have a mix we can listen to for specific ideas though?

AntonHunter said:

Different problem (specifically for live recordings with lots of spill), but I liked the idea behind this approach. Basically if you're struggling to get the vocals to sit into a busy mix, start with the vocals and fit everything else around them?

I like this idea - thanks.

ewal

AntonHunter said:

Do you have a mix we can listen to for specific ideas though?

Nothing recent. Actually maybe I do. Will dig it out of my bin ;-)

Winny_Pooh

How about a sound clip? 

ewal

Watch this space. Seeing if I can find something I'm not mortally embarrassed by.

flying_pie

Performance matters the most. End of.

I do really stupid covers from time to time and often I'm singing in a really uncomfortable style as part of the joke. I struggle like crazy when I'm dying of embarrassment listening to things that are a push to carry off or just cringe worthy in execution, but if it's a vocal I feel good about it's so much easier for it to feel comfortable in the mix to start with. Nailing the take is by far the most important thing.

After that you can worry about the technical details of the recording, and mixing process but compression and automation are definitely the most important as far as that goes

Littlejonny

I understand the problems you're having - been there. 

Ultimately, you might have a voice that's not particularly suited to the style of the backing. Also, check you are singing in the best key. It's my view that when you're doing songs, the vocal is King, so the vocalist chooses the key for the timbre they want in their voice. If they want to belt, then that's going to be a higher key, if they want to relax and go deeper, then again, you have to adjust your parts. 

You could try fairly heavy parallel compression too, and add plenty of presence with an EQ plugin. Always try and use your ears rather than your eyes.

ewal

Turns out I couldn't find anything I'm not totally embarrassed by, but thought I'd share something anyway. This was a one take recording this evening, no tuning fixes, but an attempt at using a DS'er, EQ, Comp, Radiator plus Delay and Reverb sends. It sounds rubbish. The backing tracks will ultimately get re-recorded too, but the basic issues I have with vocals are all there.

https://drive.google.com/file/d/1JjkqOXljZW739Q7ARxYqXUdNhfYtfsfo/view?usp=sharing

Winny_Pooh

Practicing singing (scales, runs, diaphragmuse and muscle control) will help articulation, even if you don't use a wide range. Best use of your mixing time really

In the link you shared the vocal level needs to be LOUDER (if you dont like it louder then see point above re practice) and then kinda wide bass cut at around 100hz by 4-6db and then try a second cut around 350hz where you can experiment with the bandwidth and level to try and bring it forward. Then try a general vocal compression preset and aim for 2-4 db gain reduction to start with, then try a short slap length delay (one repeat, highs pulled down) around 50-100ms until the vocal feels bigger then pull it down so it's less obvious. Then try a Left Right super short widening delay (diff delay times).  

You might want to cut around 1-2khz on the guitars to make space for the vocal detail in that range. 

ewal

Gosh - thanks. I can sing better than that - honest! Feedback appreciated.

theatreanchor

If you use less reverb and delay, your mixes will sound far punchier. 

ewal

Yes. But reverb and decay is a feature of the musical style I've got in mind. That particular snippet is not a good example though (wrong instrumentation etc) - in this case I agree that it'd probably be better with less reverb and decay.

When I listen to songs as reference tracks, I actually wonder if part of my problem is I don't use enough reverb and delay  Although it'll be more that I don't use the effects in the right way for the style I'm aiming for.

Anyway - I've listened to this particular song too many times now in my attempts to add a vocal. It's for this reason I'm not sure I'll ever be able to record and mix myself singing. I lose all connection with the song in the process. Think I'd be better getting someone else to sing.

Littlejonny

ewal said:

Turns out I couldn't find anything I'm not totally embarrassed by, but thought I'd share something anyway. This was a one take recording this evening, no tuning fixes, but an attempt at using a DS'er, EQ, Comp, Radiator plus Delay and Reverb sends. It sounds rubbish. The backing tracks will ultimately get re-recorded too, but the basic issues I have with vocals are all there.

https://drive.google.com/file/d/1JjkqOXljZW739Q7ARxYqXUdNhfYtfsfo/view?usp=sharing

This is great! Voice sounds pretty good too. I’d say you’re probably going to need a bit more top end on the vocals to help cut through the wall of guitar. Also, you can probably be quite heavy with the compression too. I would also turn the vocals up a couple of dB 

Littlejonny

Littlejonny said:

ewal said:

Turns out I couldn't find anything I'm not totally embarrassed by, but thought I'd share something anyway. This was a one take recording this evening, no tuning fixes, but an attempt at using a DS'er, EQ, Comp, Radiator plus Delay and Reverb sends. It sounds rubbish. The backing tracks will ultimately get re-recorded too, but the basic issues I have with vocals are all there.

https://drive.google.com/file/d/1JjkqOXljZW739Q7ARxYqXUdNhfYtfsfo/view?usp=sharing

This is great! Voice sounds pretty good too. I’d say you’re probably going to need a bit more top end on the vocals to help cut through the wall of guitar. Also, you can probably be quite heavy with the compression too. I would also turn the vocals up a couple of dB 

Actually on second listen, ignore my comments appart from turning it up.

one thing you could try is some slap back delay to give it some depth

ewal

Thanks. All useful suggestions.

I do intend to record vocal properly and also re-visit instrumentation - I think I could do the layering of guitar, drones, etc better now.

Also might experiment with double-tracking vocal. Again when listening to reference tracks, it seems to be a common approach.