What Makes A Great Vocal Mix? 

Some additional thoughts/corrections: 1) In Bartender, I'm pretty sure T-Pain is actually singing roughly those notes anyway, so even there the register has a lot to do with his delivery. There's just a layer of autotune on top of it for aesthetics. (Also I notated the rhythm wrong on the trill I think. Whoops!) 2) The soundbox model was described independently by both Moore and Danielsen around the same time in 1993, but throughout the video I mostly refer to it as Moore's. This is not to diminish Danielsen's contribution: It's simply because Moore is the one who builds on it to connect with Hall's model of proxemics, and is thus (to the best of my knowledge) more relevant to this specific discussion. The model belongs to both of them. 3) It may seem odd to describe Witherspoon's vocals as being in personal space when there's so clearly a prominent guitar driving behind it, but remember, proxemic zones vary by culture, and I'd argue that carries over metaphorically to musical zones as well. That sort of accompaniment is very normative in metal, and as someone familiar with the conventions of that genre, his voice feels like it's very close in that line. This is kinda the issue (and the benefit) of such a subjective taxonomy. 4) I should note that Duguay specifically looked at songs featuring a male rapper and a female singer, and her analysis included reflections on the gendered implications of that set-up. And I agree with her conclusions on that front, but I chose to leave that detail out of the video for a couple reasons. First, while it's harder to find examples of a female rapper/male singer collab than the other way around, the ones I looked at (like Beauty And The Beat, by Justin Bieber featuring Nicki Minaj) still mostly follow that same pattern. And second, Kimya Dawson is nonbinary but wasn't out as such in 2013 when Hokey Fright was released, so it's not clear to me exactly how to apply that framework in the particular context of the songs I was discussing. It's doable, certainly, but would require a lengthier digression than I felt was appropriate. 5) I'm honestly not convinced that range belongs in a discussion of vocal placement, as separate from vocal delivery. In the soundbox model it makes sense, in that it's being situated on the spectrum of registers within a full mix, but if we're viewing the voice as a fully separate object like Duguay's system mostly does, it's not clear to me that pitch concerns shouldn't be left in the same bucket as, say, phrasing or tone. It's clearly an important question, I just don't know if it's the _same_ important question that the rest of the model is answering. But it's not my model, and including range helps it serve the function Duguay designed it for, so I also don't think including it is wrong. 6) Confession: The loop in TV On 10 _is_ about 6dB louder than the loop in Teleprompters (comparing isolated clips of the original album tracks), but it's also much more bass-heavy, and because of how human hearing works, bassier things sound quieter, so I did bump it up a couple extra decibels so that it would be clear from a 4-second clip that the thing I was saying about it was correct. The higher decibels would still create the sonic environment I described even though the apparent perceptual volume of the track may not be as loud, but I fudged the demo a little bit in order to make the relevant details as clear as possible to a casual listener. 7) To be fair to myself, it's not like I can't hear effects at all. Most of the stuff Mike told me, I was already aware of, I just wanted to get outside confirmation. The big question I had was the compression: I could tell that something was going on with Aesop's voice, and I thought compression was the most likely candidate, but I wasn't sure and since it was a key component of my analysis I thought it was worth double-checking. But I think it's good to admit your limitations so I wanted to be clear that I did run this by Mike, and I also wanted to give him the credit he deserved for being down to lend me his expertise! Check his stuff out: www.youtube.com/@MichaelWuerth 8) Duguay doesn't actually count layering of different voices, which is fair, but I wanted to talk about it in these songs because I think, in Aesop's verses, Dawson's voice fills a similar role to what one might typically do with a doubled rapper, so I adapted the model to fit my needs. 9) Duguay's model includes 5-point scales for most of the parameters, but I didn't bother with those for a couple reasons. First, they're non-linear: They take five common scenarios in popular music and rank them, so going up one point doesn't directly mean getting more of the same thing. (For example, in the Environment parameter, E3 means dry echo, while E4 means significant reverb.) There's a progression to them, but it would require significant additional context to explain. Second, they're not super necessary when you're just comparing two tracks. I think the point of these ratings is to allow her to quickly compare across entire corpuses, and at that point standardization is really valuable, but for the purposes of the analysis I was doing, it felt like a distraction. I see the point of it, but the details of the rating scale are only gonna be relevant to actual theorists trying to employ the model for broad corpus analysis, and that's not the target audience of this video.

Fun psychoacoustic note: When a sounds position is somewhat ambiguous, we actually do hear higher sounds as above us and lower sounds as below us!

True story: I have 3 monitors and this video was up on my left monitor while I was working on the center monitor. When you panned your mic to the right I looked to my right monitor instinctively. 😄

Hey 12tone, I just wanted to say while there aren’t that many comments that I really appreciate what you’re doing here. You explain things so well, and I love listening to you talk. You’ve gotten me through some stuff, and I wanna say thank you. You’re amazing man keep doing what you’re doing.

This video gave me a lot of perspective on how I mix my own vocals. As a singer with a pretty soft voice, my vocals often end up in the “personal space”, which ends up more effective with more personal/vulnerable themes and content. Now i see why those songs of mine do better, and now I can do it with more intention 😊

Your illustrations are always really nice, but I especially loved the use of x/logx to represent "useful approximations"

Love to see Aesop get some love anywhere. The man is a lyrical genius, a true word smith full of deep introspection and beautiful cadence.

Not me learning for the first time that the melody from Gangsta's Paradise was actually written by Stevie Wonder

YES! So glad people are drawing attention to the importance of post-production, so many people see it as a bland, corrective process but its a really fun creative experience to mix a song you can really control how a song is experienced!

I just wanna say, I would have clicked on this EVEN FASTER if you mentioned The Uncluded in the title or thumbnail. LOVE Aesop, love that you covered this project on your channel.

Sort of tangential, but Joe Meek recorded in a studio that was just his apartment and the stairs basically. I don't have the quote off hand, but to paraphrase it "the singer was in the living room, the guitar was in the bathroom, the saxophone was in the kitchen, and the brass was down the stairs into the next floor entirely". (Songs in the Key of Z by Irwin Chusid)

There's a really good hip-hop track from 1997 called "Latyrx" (by the hip-hop duo who are also called Latyrx) where they rap at the same time and it's mixed so one MC is in your right ear and the other in your left. And at times the rhyme schemes overlap/interact as well, really cool.

Unlike mayonnaise, which is NOT an instrument.

Hokey Fright is one of my all-time favorite albums and I'm so glad you did a video on two of its songs!

Oh hell yes! I used to be an audio tech, and mixing, editing, signal-processing and filtering, the original sounds get changed a whole lot.

transients and frequencies come before reverb, a sound without the detail of transients and without high frequencies the brain perceives first than reverb as far away. very usefull skill in mixing music to put a element far away is take away it's transients and high frequencies first, after that u use reverb.

I always enjoy your videos! Keep on rocking and making amazing videos!

the way you explain everything sounds so dense with experience and understanding. Hope I can clear my thoughts and speak about music like this one day without reading off a teleprompter.

Absolutely love the love for the Uncluded. More weird niche content about weird niche music plz.

Topic already went so hard, but you had to go get the absolute masterpiece off of the shelf to demonstrate. This is twice what I asked for, thank you.

This is a fun video that discussed a side of music theory I don't think about often. Great work!

I was messing around with a layered vocal the other day. I was just using it to fill out my voice without relying on echo and reverb. I was planning on going through later and mix out all but the very best takes, but I wasn't ready for that yet. I was playing it back en masse, and it was sounding pretty muddy, so I started panning the voices. Some I left down the middle, then some left and right, but I found that even going farther than that, setting them at different panning levels. It started to sound much more like a chorus instead of just a mess. I'd really like to take it farther and play around with surround sound, but that's beyond what my setup can handle. I'm a big fan of shape note choral singing. Some of its the melodies, but I think part of its power, when it's performed live, is that all the choral sections are facing each other in the round. It creates a very different sonic experience than a choir that's performing for an audience.

13:53 is that Jak from Jak 2? When you said "this time we're doing it a little differently" ?? That's so obscure if so. I love it!

While panning is one way to move the position a sound sounds like it's coming from, it's not the only way. You can also play the sound through both sides but adjust the phase of one of them, or, more simply, put one side on a slight (very slight) delay. You don't want enough delay to be outright noticeable, but a very small amount of delay will adjust the relative phases of the sounds and this gives our brain information about where a sound is coming from. If your left channel is slightly delayed (by like 0.6 ms), then it will sound a bit like it's coming from the right even though the volume on both channels is the same. Less delay makes it more in front of you. It's hard to experiment with, but one artist that did was Trent Reznor of Nine Inch Nails on the EP Broken. I'm not sure exactly what he did on that EP, but he has said he messed with the phase of various instruments on that EP. When you listen with headphones, everything is kind of weird sounding as a result. Unfortunately, doing this has the side effect that you pretty much have to listen with stereo headphones or parts of the mix will randomly cancel out.

Hey 12 tone could you do a video series explaining the different albums and songs of radio head examples being: how to disappear completely, creep, kid amnesia, etc. They are one of my favorite bands that have so much hidden meaning in the albums and songs.

I usually don't pay much attention to the drawing and writing in the videos, but I really appreciate you saying "useful approximations" and writing down the first good approximation to the prime counting function. Not too obscure but not too well known, it's a good choice

In “Heroes” by David Bowie, co-produced by Bowie and Tony Visconti, the presence of the vocals changes very dramatically throughout the song. Bowie goes gradually from whispering closely to screaming with reverb as the song progresses. The Wikipedia article describes how this was achieved, as well as a RU-vid video of Visconti himself demonstrating it with isolated vocals from the actual recording.

Liam Gallaghers vocal mix on definitely maybe is biblical . I recorded a version of where is my mind years ago and the singer covered himself in a blanket and sang into a cardboard toilet roll, I’m not joking - it sounded great . We also recorded guitar tracks in a toilet once, sounded great. It’s mad how sound waves bounce off each other

I've been following for a while and love your work, but this one... you know those moments when something just *clicks* for you? I do some podcast production, and am looking into audiobook production. This video was *fascinating* because I'm pretty much self-taught in those spaces, but reveals a framework to things that - on some level - I instinctively grasp at. Since those areas are *heavily* (if not exclusively) vocal, the more complex model was a wonderful insight. A couple minutes in I switched from my computer speakers to my headphones so I could more closely listen to your examples and what was going on. Well done. I'd love to see more

The Edward Hall model seems very useful but he forgot one zone: shoegaze space. The space where you can hear that a voice is present in the mix but you have to do your absolute best to make out 10% of the words that are being sung.

I remember the first time I noticed the panning in Queen's Now I'm Here when I was 7 or 8 years old and being completely amazed by it, but then I didn't know what mixing is lol

I very much enjoy your channel even though when you do a deep dive into theory it often gets past me I still manage to learn something. BTW, do you singlehandedly manage to keep the staff paper industry solvent ?

A CrossCode reference in a 12tone video? A man after my own heart...

3:40 This was the most exciting part for me as a music creator, the concept of Prominence and that is has four components, Volume, Reverb/Echo, Timbre, and Perceived Volume. each of which affects the others but is also independent. Slightly baffled that there was no visual breakdown or highlighted text. Feels like you didn't think this was important in a vocal mix. (And ngl every music example you used was pure ugh.)

I totally agree with the point here. The preconception that "a recording should be an accurate replica of the live performance" i not only think its off base, but flat out backwards. The live performance should be like a picture of a picture, a representation or alternate take of the perfect version of the tune (as represented in the studio ideally) Like i love live improv and experimenting but to do so tasefully is an art in itself that takes into account the fact that the people who come to listen expect to hear certain things. A good example is The Greatful Dead. they experiment everywhere but always "sound like themselves".

An informative and interesting video as always, thanks for that! Though having some experience with mixing and acoustics I have a consideration: with any stereo set, or even mono speaker, headphones and whatnot, you can very well convey clear distance information using psychoacoustic. The clearest information comes from the ratio of early reflections to late reflections, with more ER giving an impression of the source being closer, and more reverb; then the delay between the direct signal and the first ER together with the ratio of direct to reverberated sound can give an impression of a bigger or smaller space. These parameters are all very easy to manipulate with most reverb processors, and they're not usually taken into consideration in busier mixes because they're not the easiest thing to program, but they're often harnessed in more sparse mixes, for example just a voice plus a guitar, and it's quite interesting to hear how that's managed in different productions. If you're looking for a quick and dirty solution there's also an old plugin called Proximity by TDL and Vladg sound which has a "Distance" knob.

10:20 I would absolutely click on a video about The Uncluded and now I really hope you do one!

When talking about vocal layering, songs by Emmy Rossum come to mind. I remember listening to her songs quite a bit to distract me from the stress of studying and teaching.

Loved the CrossCode reference! :D 'Hi!'

AAAAH got my man Aes on 12 tone!!!

I'm so glad you did a video on this. In electronic music scenes, the actual notes you use are maybe 10% of the total expression of the music. There is an entire language and spectrum of emotional expression that comes from how the sounds are designed and mixed in the song. You could probably tell an entire narrative in an electronic song with one note.

The entire anthology of Rusty Lake is unique in the fact you can kinda play the games in any order and get the story in much the same way as if you played them in order of release or chronological order, which is a stretch to say since the game explores time travel as well, which is officially not a spoiler because Mr Owl literally tells you at the end there.

It is always the right time for Kimya Dawson

tbh i haven't watched this through yet but if you want a masterclass in vocal mixing you NEED to check out hiatus kaiyote's song "red room". its one of the best engineered songs ive ever heard, it sounds sooooo good

How on earth did you include a Feynman diagram into that??? I just love those little details that only 0.5% of your viewers get.

This was a surreal episode for me. When you did the comparisons of samples, I could easily hear the differences in the mixing and performance, but not once did that translate to the difference in feeling that you described. Samples didn't feel closer or further or more or less personal or more courageous or earnest or fearful. I think a lot of the difficulty I'm having with learning music comes from my experience of music being incredibly different from that of the rest of the western world. My experience of strong and weak beats is often backwards from other people. My experience of strong and weak harmonic motion is often backwards. I rarely feel a tonic. I don't feel one underlying pulse throughout a song -- unless the pulse is played.

Somehow iissed this one when i came put which is a huge shame because these are always my favorite kinds of 12tone essays. I hope you get patronage to do more voicework!

12tone, I know you already did a video on Talking Head’s Psycho Killer, but I was listening to Once in a Lifetime and realized that there’s only 2 words that rhyme that aren’t the same word in the entire song. I’d love to see a video on how it’s still such a great song without having rhyming verses.

Hi 12tone, I found your channel a few months ago, and your metal analysis videos are really cool. I would like to see you do more Metallica songs and some Megadeth. Specifically Holy Wars, as the song structure is really weird with that one.

As an exclusively right earbud user while working and listening, I will never know what happens when a mix is panned left

LET'S GO I NEED HELP WITH THIS

This super interesting just a bit niche thank you for your hard wrok

Absolutely. That's why it's so important to work with a good producer and recording engineer. Also, Aesop Rock is the 🐐

Nitpick, but I can't let this pass: In Wuthering Heights, Kate Bush doesn't sound close at all--she sounds like she's out on the wily, windy moors.

i think give peace a chance by john lennon and yoko ono is an example of public space

What You Need by The Weeknd is a great example of a deliberate atypical vocal mix.

This is a great overview of vocal mixing technique from a theoretical and somewhat scientific standpoint. However I can pretty much guarantee that the mix engineer was only thinking about whether whatever they were doing felt right for the song. All the theoretical knowledge in this video is in the back of an engineers mind while mixing if they consider it at all. A good mixer knows all of this instinctively and rarely if ever does this level of analysis play out in the studio.

I never thought I’d see a Tobacco contemporary & collaborator on this channel

Me casually listening to this on the drive home: 🙂 8:21 Me: 😳 "Did they just f**king reference SEVENDUST?!?! 🤯 RIGHT ON!!!🤘😀🤘

My favorite song that hard pans a lot is Such Great Heights by Postal Service. I can’t wait to see them play all of Give Up and my favorite band Death Cab for Cutie play all of Transatlanticism.

Could you do one on a elliott smith song? That could be interesting

Hey! I wonder how language affects this. People often say that some languages 'sound more romantic' or similar, and there might be some music theoretical basis for that? Like American English happens in a different part of the mouth than e.g. Scottish English or French, so the sounds also generally sound flatter and higher while Scottish sounds fuller and lower. Is that a significant thing? And could this explain why some genres are much larger in e.g. Scandinavia? (Like I could imagine that the American English accent is bad for metal, while Finnish could be better -- even if the Finnish person is speaking English.)

I mean, what makes good mix overall

7:54 LEA CROSSCODE IN THE WILD????????

Ya made my ears pop

1:50 i'm waiting for the comments to roll in from people who are now realizing they have their headphones or speakers set up backwards lol

Please do a RU-vid channel on album sequencing 12tone.

That bit about breath intake reminded me of why Kaine Salvation from the NieR games hits so hard, the mix picks up on Emi Evans’ full breath before each phrase, giving it a more vulnerable mood that really defines it

As an ammeter producer, thank you.

I feel like you were trying to demonstrate panning, but I listen to RU-vid in mono.

3:00 is a much better explanation of why I think large parts of _Hamilton_ ultimately don't work: because no one outside LMM can be understood as Hamilton. Saw the tour recently and whole that night's Hamilton was great, he was very difficult to hear at times because he was a more traditional, round-tone baritone. LMM's speaking voice is brighter and possibly pitched higher than many others performing that rule, and his voice is what the orchestrations were built around; without having the richer overtones and resonance to cut over the constant cello and similarly low instruments, Hamilton's voice gets lost in the music.

Why in the clips from the intro you are righhanded and then you write with your left hand?!

So is notation software... Finale is essentially an instrument.

Not that it's relevant to the topic of this video, but the other thing intimate space is used for (besides sex and violence) is secrets.

Should I understand "Delay" every time you say "echo"? I think so. Great video

Would "Heroes" by David Bowie be a decent example of changes from intimate space to personal space to public space?

hi, i know im not a patreon cause i'm still not financially stable but i would just like to give the suggestion of "Belinda Carlisle - Leave A Light On", i don't know why but it's not a BIG song but it's still pretty good and it hits different. hope you are doing well :)

Doubled, compressed and saturated, maybe even with a phaser; i like my vocals as if they r comin through a megaphone 📣

can u do an understanding music on sweet child o mine guns n roses pls

The crosscode reference lol

You totally forgot to mention one of the most obvious reasons someone might stand close to you and whisper, they're telling you a secret. Intimate space is also used a lot in music for moments when it feels like the artist is confessing something, something hard to say.

Could you please do an analysis of Kashmir by Led Zeppelin?

You should do “understanding deftones” 🙂

very cool for kimya dawson to get time on this channel

Can you let me know when Nebula gets a comments section so I can finally watch a video on it? 😆

I will click on EVERY video about The Uncluded. Please do that. I’ll send them to all my friends and also random strangers

King Tubby approves.

Yeah! Yes! Dood. Do more videos about mix theory. Change your channel name to 20kHz pleeease

So, if personal space often includes sounds like breathing, then where does "Chocolate Rain" fit into all of this? "I move away from the mic to breathe in."

Ironically I would've clicked on this video earlier if I knew it was about some of Aesop Rock tracks! 😂

interesting way to learn that my speakers are placed on the wrong side of my monitors

Echo The Dolphin to represent echo / reverb!

Uncle iro !

another technique for projecting public space - speaker is talking through a bullhorrn.

Hi Ren?

Heyyyyyyyy Kirby….

One other parameter is … does it sound cool

Feel the same of Joji, When???????????????????????

First

This a gross over analyzation. When mixing, we just do what "sounds right" for each song/mix. This is like when music theorists try to justify a certain chord or note choice. The reason is quite simple: it gives us the sound we want/like.