Hi guys,
So a few days ago i had the privilege to mix a vocal take that was recorded through a C414-1073-CL1B (ive never mixed anything that was recorded with such good gear haha)
and the one thing i noticed immediately that was different from other vocals i mixed, was that the vocal had a really nice crispy clean high-end - but it wasnt 'essy' at all! i even boosted abit of 8k with an SSL plugin and still didnt need to use a de-sser! i was really shocked. most of the times i need to de-ess vocals immediately since i mostly work with mic-to-audio-interface budget rappers.
so, obviously it depends on the vocalist and matching the right mic for him and stuff,
but im also pretty sure that its thanks to the great vocal chain combo -
TLDR:
so my question is
1) How would you go about getting a nice clean high end vocal that doesnt need de-essing - with a budget pre amp and compressor? (hardware budget pre amp & compressor recommendations)
2) what are some de-essing tricks you use to de-ess while trying to preserve the high end detail in the vocal? (with hardware / plugins)

 

The best way to achieve a signal that does not need de-essing is by de-essing the signal because afterwards you do not need it anymore . Why is it relevant if you need de-essing or not? You do it if you need it. If you do not have good recording equipment you have to do more processing to get decent results.

This depends on the behavior of the frequencies. If the frequencies are very harsh than you should try notch filtering because at this moment we are trying to get rid of errors of the behaviors of bad recording equipment. If you have too much high frequencies you can try an high cut and if you have to much "ssssss" going on in specific situations of your signal you can try de-essing.
You can not say that everything that has to do with "my high frequencies sound good" is because I've only used my de-esser properly.

The "trick" is to use the correct tools in the right moment.

 

TLDR: do a Cut at 1k hz. or around that.
if your ears are still not happy, try around 3k and 4k also commenly be a problem if too many tracks pileup oneachother around those freq's.

 

What i mean is that the recording is rich in high end - yet it doesnt sound harsh, therefore i dont need to de ess it,
Im not used to it, usually i get harsh sounding vocal from consumer grade audio interfaces and if i boost some high end (which i always do) i need to de ess, but this time the vocal was bright enough and it wasnt harsh
So im asking if anyone that had this experience and could reccomend another (cheaper) way to achieve this out of the box in the recording process

 

I had an AKG 414EB with a brass capsule as well as a Rode Classic as my go to vocal mics depending on the singer and neither exhibited any sibilence. These days my Gefell mics do an equally good job. In my experience the microphone makes the greatest impact, as even on a sibilent singer a fine mic can be forgiving. Choose your microphone by auditioning it at home in your room or through direct experience out in the world. Next up, the preamp can have a magnifying effect on the sssss thing if it is at all present. A pre that can saturate will lob off those sharp edges of the ssss transient and tame the snake. FabFilter's dess plugin is great, the only de-sser software that I have used.

Mic placement can help a great deal if the singer is sibilent. Being a little off access sometimes can make a world of difference.

 

deess, excite, deess again, eq, then deess again and so on. You may need several n-band compressors, each band coping another nasty frequency (from 2khz on).
What you don't need is 'that one magic plugin'.
absolutely no rocket science.

 

I just de-ess and/or manually volume automate. I also use Pro-DS.
The ssh and esses are usually around 6 to 9k. The high end "air" is above 10k. That's where the real yummy shimmer is.
I also use the Dolby exciter trick to compress the upper ranges. Great trick (to make great sounding vocals) which not many know about.

Sibilance is not "catched" by compressors, as they barely have any energy. So people who compress their vocalists/voices tend to have the sibilance quite strong (when using LDC mics, etc). Hence why dynamic mics (with not as fast highs) "sound better" when being compressed quite heavily. The end result is like a neat compromise.
This is why some female vocalists can sound great one dynamic mics, and vice versa. Especially when added saturation, which also enhances upper harmonics.
I've used Manley Reference mic on a female vocalist and it sounded like shit, but the mic is expensive af. Hence you you should match the mic to the vocalist.

The 414 is a "neutral" mic which I really like (one of my favorites). I've used both the older (silver) version as well as the newer brass ones.

 

Last edited: May 20, 2019

what does De-ess mean?

 

Automation!

 

First of all I always make sure that any high frequency boosts are done before slamming it into any compressor,
After pitch correcting and noise gating to be exact.

then, the trick for me is, keep switching out eq's for the high shelf boost.
tdr slick eq does a decent job at that. Slate vmr's, eiosis air eq, any ssl emulation are worth a shot.

If it still needs de-essing, fabfilter would be my favorite weapon of choice.

 

not to discuss semantics, but isn't that strickly speaking a double layered expander on the high frequency's?

 

vocals need midrange not crispy high end. a nice 1073 and tape plugin will round the edges nicely

 

You could try this (click this link).

You can use it as you need and want to.
If you're really pleased about it, and if you can afford to, you can Patreon it (click).

Chris Johnson from Airwindows really deserve to be Patreoned.
He is giving tons of brilliant and useful an original unique audio plugins to the community, whether you have money or you're broke.
A real blessing for all (from broke musicians to high end audio engineers).

 

Last edited: May 20, 2019

Essentially with a pretty good mic (I have condenser mics and a ribbon mic which doesn't need de-essing, while other condenser mics I have are needing de-essing).

 

Last edited: May 20, 2019

Gain automation of Ssss, Shhh etc. until it is enough. Slight dynamic eq or good deesser for resonance whistling sss, to surgically reduce earfatigue resonances.
Then work with high-end as you want. Add exciters/enhancers/eqs/trebleboosters etc.
TDR deedger as well for killing sharpness.

 

It's a process to lower or suppress annoying disgracious sibilances on Sss and Shh sounds within vocal recordings ; generally it's a frequency dependent compression, a compression at the frequencies where these sibilances are occuring.

These sibilances when happening can ruin a vocal take and make it difficult to mix, if left as they stand.

 

Last edited: May 20, 2019

Been there. +1 For mic placement, it's the source after all. Just be aware of how you're changing the sound when placing the mic. Then get surgical with an EQ. Start from 1kHz at least and go up, notching out frequencies, adjusting previous bands to preserve a natural sound. Sometimes starting from 20kHz and going down works too. Graphic EQs and harmonic EQs work well for this. I like the API 560 but you could go crazy and use a 32 band. A nice trick that works well when working on high frequencies, I saw a long time ago, I can't remember from who, was to put a low cut on at a fairly high frequency and EQ the highs. Makes it easier to hone in on high frequencies. Anyway, after you EQ the highs, find the telephonic frequency which is usually around 2kHz and cut it. Sometimes removing even a little bit of that telephonic frequency may ruin the recording so try a dynamic EQ or a multiband comp for a gentler approach. While you're at it you could move some of the high frequencies from the first EQ to the dynamic EQ/multiband comp. Then you could try overall high frequency boosting. Or! If you want to go further after that, get two analogue (emulation) EQs with different flavours and boost high frequencies in one and cut in the other. This will drastically change the sound of your high end. This trick is a good place to either audition a sound to work towards or as a general starting point. McDSPs 6050 Ultimate Channel Strip does this extremely well with it's array of vintage EQs and Comps and other tools. Another step further would be something like cutting high frequencies before a comp then adding those back after or even adding high frequencies before the comp to achieve a sort of manual deessing. Other things I could think of is tape or tube emulation or saturation in general like someone above mentioned. Definitely experiment with this process, rearranging as you see fit. If you've got money to spend then the ART Tube Studio MP V3, despite being a starved plate design is a good place to start at around $80 USD. It'll do some warming for cheap and you won't find any gear worth more in value for what it does until you spend maybe double that. I guess you could get an EQ or comp around that price too so you could at least have something to add to the chain if the V3 doesn't seem worth it. Glad to help. Or you could skip all the above and buy yourself a Neve console. I'll be here all week.


EDIT: Gain automation is a given. If it's not part of your vocal mixing process then it should be. Both before and after effects. There's magic hidden to be had from vocal automation and automation in general.

 

Last edited: May 20, 2019

This guy knows what I'm talking about :3

 

Personally, I use a combination De-essing via subtractive EQ (very minimal cut; -1 to -2 db with wide Q), then I use a Dynamic EQ to address them further. The order of operation defers at times, but the mentality is the same. In short, contrary to belief, "esses" and "shzz" are important as they give definition and clarity to a vocal phrase, however, too much of them will make your ears bleed.

Generally, one should not try to eliminate them, but instead have them under control via a dynamic EQ plugin or a multi-band compressor targeting the abusing frequencies.

 

Eq---compress(saturate)----deess----then eq the top end.