Mixing lower range vocals

[Bassaga]

I'll be recording and mixing my first album in the next few weeks. Pretty excited, and learning a lot of things pretty successfully. But I get hung up on the bit where everybody says, "When you're mixing vocals, x y and z are bad frequencies, and a b c are good frequencies."

I can see where there's a defined sonic space in the band for vox, and where the majority of hit singers are in that range. But my voice is decidedly lower. I'll do some high singing, but a lot of low singing as well, sometimes in the same song.

As a drop of one octave cuts all the numbers in half, I just have a hard time believing that adjusting the same few frequencies will have the desired effect on all vocals the way people say.

I know the ideal would be to just try it all out different ways and figure it out myself, and I'll give that a shot. But like I said, this is my first album. I've never mixed anything before, and frankly my ear just isn't that advanced yet. So I thought I'd ask you folks. Do all vocals respond well to treatment at the standard set of frequencies, or do unusually high or low vocals require different treatment?

 

[EvilJ]

In my experience, any mixing "formula" is rarely a good idea, and it will usually lead to a below average or worse mix.
Every song, singer, instrument, microphone etc. has a unique effect and should be mixed as such. Unless you are always working with the 'same' source material, i.e. same band, mics, instruments, preamps etc.

Jesse

 

[pklima]

High vocals are probably easier to mix, but you do not want to cut the recommended frequencies in half then mix... if anything else, be even more aggressive about cutting the lows on any low backing vocals or extra layers. And you might need to boost more upper mids/cut mids in the instruments to get enough clarity. Or ideally arrange the instruments in a way that leaves room for the vocals above the kick and bass, but if you're recording and mixing in a few weeks, might be a bit late to be redoing arrangements too much.

Don't worry too much about cutting the lows, or if vocals sound thin soloed. That's normal. The whole needs to sound good - which implies the total balance of all frequencies should be roughly the same whether singing high or low.

What sort of style and instrumentation are we talking about here, though?

 

[And I]

There are no "magic frequencies" that should always be adjusted a certain way in every mix. "Everybody" who says that is nobody you should be listening to.

 

[Tosh]

I like to employ frequency analysis on a track before I do anything with the eq, then make my decisions based on how it sounds. Vocals, drums, bass, all of it. The only instrument that I *almost* never EQ is my keyboard (because I already eq'd it beforehand).

Though I can set up a templat to work from, it's never truly right. No formula, just a guideline

 

[Anthony Buckeridge]

With all due respect.

I believe you are thinking about this matter completely the wrong way round.

Instead of thinking about what frequencies to adjust after the voice has been recorded to obtain the desired sound.

Think about utilising the optimal microphone whose sonic signature will enable the characteristics of the voice to best convey the emotions aroused by the song.

If you have a great sounding voice to record, why not capture its performance, just exactly as great as it is, without the need for any further post recording “enhancement”?


If the voice isn’t that great, why bother to record it at all?


I would be looking to use a mic with a slow, gradual, slope off the lower frequencies from 1,000 Hz all the way to the bottom of is range.

In recordings, its easy for the lower middle range to become thick and cloying or muddy. But this area is often boosted with equalisation which makes the problem worse.

By utilising a mic that rolls off those lower frequencies, you preserve and enhance the voices clarity and conveyance of emotion and separate it sonically from other elements of the mix, that might otherwise compete for similar space.

If you click on “data and diagrams” and scroll down you will see this mic rolls off from 5 kHz all the way down.

TLM 49 - Studio Microphone (neumann.com)

Here’s another popular mic. See the effect of using its Low Cut Switch.

U 87 Ai - Studio Microphone (neumann.com)


By using the right mic of excellence and the right placement of the voice in relationship to it position in front of the mic, it should be possible to get the sound you want virtually “record ready”, sounding like the finished article.

You may want to use a little compression and reverb but it should be possible to get really close to the finished sound without adjusting the equalisation at all whilst recording, even if a touch is needed during the final mixing stage.

I’ve worked on a few projects that have involve singers with low voices, and suppose the late Johnny Cash would be one of those with the lowest sounding voice. But I would strongly encourage you to leave the equalisation flat in the first instance and focus your attention on using a mic with the type of characteristics I have described.

I think that approach will be most likely to provide you with the optimal result.


People often appear to think that boosting frequencies in a particular range is the way to obtain the sound they want, and sometimes it is.

But where lower frequencies are involved (bass instruments apart) it is often preferable to use minus equalisation (the microphone above do that for you) to preserve clarity.

Take a piano for instance that extends across a very wide compass of frequencies. If you boosted the lower end with equalisation it would begin to sound a bit like a bass, and directly compete with it in the mix.

But if you sloped off its bass frequencies from 100 Hz gradually downward, suddenly there is clarity and that hallmark zing of a genuine piano bass sound; without the cloying muddiness that accompanies boosted frequencies where such boosting just makes everything too thick, giving an unclear sound.


As the sum total of voices with their reverb and instruments add up into the mix that whole area of upper and lower mid frequencies and bass frequencies just become to saturated in plus equalisation.

One thing I learnt from Gus Dudgeon was that as most instruments, particularly if an orchestra is involved, occupy the mid-range of frequencies. Rather than boosting those areas to help bring them out, it is usually better to use minus equalisation to provide clarity. If you listen to Elton’s early hit records you will readily perceive the clarity this provided.

Furthermore, if my memory serves me correctly, it is long been virtually a standard operating procedure at Abbey Road recording studios. To send signal to the reverb chamber, applying a high pass filter at 600Hz and a low pass filter at 10kHz. This cuts out the muddiness of low middle and bass frequencies along with the top end splishy spashiness that can otherwise spoil a great recorded sound.

The returning sound from the reverb chamber is much cleaner and clearer and minus the unwanted artefacts that using effects can often add to a great recorded sound.

Try and get the sound you want from the voice at source, and capture it with the optimal mic used in the optimal manner without using any equalisation at all whilst tracking.

I learnt the above from the late Al Schmitt and it seemed to serve him well.

 

[Nephilymbass]

I'll be recording and mixing my first album in the next few weeks. Pretty excited, and learning a lot of things pretty successfully. But I get hung up on the bit where everybody says, "When you're mixing vocals, x y and z are bad frequencies, and a b c are good frequencies."

I can see where there's a defined sonic space in the band for vox, and where the majority of hit singers are in that range. But my voice is decidedly lower. I'll do some high singing, but a lot of low singing as well, sometimes in the same song.

As a drop of one octave cuts all the numbers in half, I just have a hard time believing that adjusting the same few frequencies will have the desired effect on all vocals the way people say.

I know the ideal would be to just try it all out different ways and figure it out myself, and I'll give that a shot. But like I said, this is my first album. I've never mixed anything before, and frankly my ear just isn't that advanced yet. So I thought I'd ask you folks. Do all vocals respond well to treatment at the standard set of frequencies, or do unusually high or low vocals require different treatment?

Mixing vocals is a tough thing. Especially if the singer is very egotistical. I wouldn’t worry that much about messing with certain frequencies ranges that are problems for some people if that likely doesn’t apply to your vocal range. I’ve experienced a lot of the bad of having big sounding vocal tracks that stomp all over the mix. Something I really noticed from listening to Rick Beato’s what makes this song great series. Is a lot of pro music male vocals are high passed and a bit more than you would think listening to the entire mix. Lot of the time when the Rock solos the vocal in those videos it almost sounds like a megaphone effect because there’s so much hi pass eq on the vocal. If there are vocals in spots with no music behind them you can put those on another track and don’t hi pass them so them so they sound natural and then when the music comes back in you can have the vocals on the hi passed track(s) so it doesn’t muddy up the mix. Or another route if you don’t want to have separate tracks is to use automation to turn on and off the eq. But yeah I’d focus more on just using a high pass instead of going after several different frequencies that might not even be an issue with your voice

 

[And I]

please disregard any "this frequency is good or bad" advice. gotta use your ears...

if this is your first mixing project, i'd recommend "Zen and the Art of Mixing". there is a good deal in there about mixing as a job which maybe you don't need, but even more good information about mixing as a process--which we all can use!

 

[David Vector]

A thing about recording a deeper baritone is that you can kind of fall in love with that "radio announcer" fullness such voices can have (especially if you use a mic like a ribbon that really lets the lower end bloom). I used to try to keep some of that thickness in the mix and found I just had to let more of it go or the vocal would never slot into the mix right.

fwiw, I've found oeksound soothe helpful with vocal mixing. As the vocal part goes up and down through pitch ranges, it can attenuate strong resonances wherever they occur, whether it's strident vowels between 1 and 3KHz or excessive low-mid bloom down lower. It cut out 90% of that manual automating of multiple EQ bands I used to have to do on vocals. (But, as with any tools like that, you don't want to overdo it or things will end up sounding overly flattened out and lifeless.)

 

[YourGuitarist]

I can see where there's a defined sonic space in the band for vox, and where the majority of hit singers are in that range. But my voice is decidedly lower. I'll do some high singing, but a lot of low singing as well, sometimes in the same song.

Do all vocals respond well to treatment at the standard set of frequencies, or do unusually high or low vocals require different treatment?

Well if you cover a lot of ground in terms of octaves in your singing, I would mix them separate from the lower pitched stuff. Automation is the name of the game here. I'm also assuming that your high notes are in head voice moreso than the chest voice with lower notes which would mean the timbre is noticeably different and would warrant different treatment between the registers.

The only "deep" singers I know are Depeche Mode, Woodkid & Johnny Cash so maybe take a listen to them for inspiration. As a somewhat dark voiced singer, I find that cutting 100-500 helps and boosting 800-1.5k helps a lot. IMO deeper singers have more presence in the 1k-3k region whereas trying to add air in the 9k+ is "futile". The truth is mixing vocals is HARD because unlike guitars and drums etc, a voice typically only has 1-2 timbres that are musical whereas I could shove a guitar into a number of frequency bands and still have it work.

Don't underestimate compression either. A lot of radio vocals are heavily limited which adds a lot of presence via sustain.