De-essing - good reference tracks?

[Zukan]

The reason some of you can't hear sibilance on current releases is because it's already been treated at the mixing stage. We producers spend as much time cleaning and optimising stems as we do adding production fairy dust.

Almost every song I have mixed, bar a very few, I have had to deal with plosives and sibilance, along with resonances and mismatched gains from poor mic techniques.

De essers nowadays are very powerful and if you really want one that alleviates time and energy in locating the offending frequencies pick a de esser with a solo function. This helps nail the exact area/s sibilance is pronounced.

More to the point, do you need a de esser to tame sibilance? Depends on the sibilance intensity and whether it resides in one or more frequencies. Clip gaining can yield great results if you have the time and is an option I often look to if I want word by word precision. There are many workarounds to taming sibilance and with the tools on offer today it's actually quite simple to hone in on and tame the offending frequency/ies.

As for reference content, won't downloadable raw vocal accapelas suffice? If that fails, then some amateur content tends to suffer from sibilance for a multitude of reasons, and worthwhile asking for submissions from forum users or social media groups.

However, it is Sunday and therefore I could be talking out of my Johnson.

[Drew Stephenson]

Sung by Elizabeth Fraser I adore her voice also like how her voice is represented in this song.
Has Synth, Guitar.
https://m.youtube.com/watch?v=HFWKJ2FUi ... Ugc2lyZW4g

Lovely song but far too spitty for me and totally distracting from the song.

Damn you Drew, now all I can hear on my last demo track is a nest of vipers. It was fine before you posted!

Slight exaggeration but I think you have marginally re-programmed my sibilance sensitivity.

Sorry! But you're in good company!

Unprocessed, part-processed, and the closest I've got to satisfactory will be coming along shortly. Your patience and indulgence is very much appreciated.

[tea for two]

Sung by Elizabeth Fraser I adore her voice also like how her voice is represented in this song.
Has Synth, Guitar.
https://m.youtube.com/watch?v=HFWKJ2FUi ... Ugc2lyZW4g

Lovely song but far too spitty for me and totally distracting from the song.

Aaaah. This gives me a trifle better appreciation.

My ears are sensitive to some spoken voices women and men, hihats rides, violins played high up, piccolo, brass, synth sounds as from Artruia pigments, granular synthesis : they can hurt my ears make my hearing uncomfortable I stop listening.
I think you have Arturia Pigments. Also you have Beyer DT880 pro on which the top end hurts my ears so my DT880 pro went to another home.

This song is fine for me listening on Shure1440 which have prominent upper mids prominent highs.

So I would suggest there's possibly something up with your hearing re: sibilance
as there's something up with my hearing for it to hurt.

Preehaps passing your vocal processing onto somebody you trust hold in high regard : there's plenty on ere foruume as we know.

[Drew Stephenson]

I think you have Arturia Pigments. Also you have Beyer DT880 pro on which the top end hurts my ears so my DT880 pro went to another home.

Nope, but I use the 770s and 990s so in the same ballpark I think.

So I would suggest there's possibly something up with your hearing re: sibilance as there's something up with my hearing for it to hurt.

Possibly, but if I can get a couple of good references then I can work around it.

Preehaps passing your vocal processing onto somebody you trust hold in high regard : there's plenty on ere foruume as we know.

Nice idea but not really practical I don't think. There's no money in this and I have too long a backlog of tracks to work through to ask anyone to do this for fun.
Besides which, it's something I think I need to be able to do.

[Drew Stephenson]

Right, raw files time:
Firstly completely unprocessed vocals from Gridlines (bit of clean gain to aid comparison): https://soundcloud.com/user-359526178/g ... al_sharing
Not crazily essey, but definitely needs work.

But the reality is that's not how it sits in a mix, so here it is with the compression added (no other processing): https://soundcloud.com/user-359526178/g ... al_sharing

Finally, for comparison, here's a different track I think I've nearly got right: https://soundcloud.com/user-359526178/n ... al_sharing

[Drew Stephenson]

PJ Harvey - Shame
Chad VanGaalen - Sara
John Cale - Sylvia Said

Just been listening to these.
The PJ Harvey is pretty good to my ears (but there's not many 's' sounds in the lyrics (plenty of 'sh' though) - either way it's going on my list), the John Cale sounds very essey to me, and the CVG is ok-ish. I'd have wanted something cleaner personally.
Obviously I understand that de-essing technology as we know it wasn't around for some of these tracks.

[Tomás Mulcahy]

Right, raw files time:
Firstly completely unprocessed vocals from Gridlines (bit of clean gain to aid comparison): https://soundcloud.com/user-359526178/g ... al_sharing

I agree with James. It's not esses, it's a set of HF resonances in the mic giving a metallic edge. It sounds like you are whispering along with yourself. As he said, people spend a fortune on certain mics for this sound. Personally I dislike it.

I remember blind listening to Lynn Fuston's Mic test CD. Twice each for male and female performances. I picked the Audio Technica AT-4033a both times because it was the only large diaphragm capacitor mic I heard that does not do that metallic thing. I used one extensively, I find it works great on a wide range of voices. I remember first trying it thinking it was too bright compared to the AKG C-414 BULS, but the vocal always sat better to my ears. Even the AT2020 has a similar thing although it is slightly more ragged sounding, if that makes sense.

https://www.soundonsound.com/reviews/3d ... re-mic-cds

[SafeandSound Mastering]

A vocal will naturally have some slightly "sticky out" consonants, this is both expected and natural.

They will stand out as the language and mouth naturally dictate.

I would say the following, and often do to clients. If the ear is being unduly pulled towards... we will say "esses" though absolutely not restricted to only "esses" then they should probably be addressed.

It is only an issue if they are out of balance. Some voices are natually more sibilant than others.

The solution is also very easy. A dedicated de-esser and/or a dynamic EQ. If unsure look on a spectrum analyser to zone into the problematic frequencies (often more than 1, sometimes up to 3) which are normally between 5-13kHz.

The overall tonality of your mix, if diverging from, let's just say 'nominal' for any given genre, may change the perception of sibilance, so bear this in mind. A mix lacking in top end, be it relevant sound sources or just tone/eq choices and that is on the warmer side may make sibilance appear worse than a mix that is slightly on the brighter side or even just "normal".

This should be quite straightforward to resolve without over thinking things.

I am sure there could be some good references of course but I am not sure they are needed. Just consdier if your ear is drawn/distracted by any excessive HF triggers throughout the vocal. That's it really.

Of course it is very helpful if your room and monitors are trustworthy.

[Drew Stephenson]

I agree with James. It's not esses, it's a set of HF resonances in the mic giving a metallic edge. It sounds like you are whispering along with yourself. As he said, people spend a fortune on certain mics for this sound. Personally I dislike it.

I remember blind listening to Lynn Fuston's Mic test CD. Twice each for male and female performances. I picked the Audio Technica AT-4033a both times because it was the only large diaphragm capacitor mic I heard that does not do that metallic thing.

Thank you, I may have to pop along to G4M in due course and see if I can try one out for future consideration (I have a big birthday coming up later this year so might be able to stretch a bit of expense).

...If the ear is being unduly pulled towards... we will say "esses" though absolutely not restricted to only "esses" then they should probably be addressed.

That's the thing, they are sticking out to me, to the point of frustration.

The solution is also very easy. A dedicated de-esser and/or a dynamic EQ. If unsure look on a spectrum analyser to zone into the problematic frequencies (often more than 1, sometimes up to 3) which are normally between 5-13kHz.

Yep, it's easy enough to identify where the problem area is. My challenge is getting the right amount of reduction to address the problem without going too far and creating more lispy artefacts.
I struggle to accurately assess this whilst mixing (though i'll always hear it a few hours later, whether over- or under-done), hence seeking out some references that I can quickly toggle between.

Of course it is very helpful if your room and monitors are trustworthy.

It was ever thus!
My room is pretty well treated and fairly consistent (with the exception of a room mode at 160hz) and my monitors are KH80s.
Neither of those things are going to change in the foreseeable future.

[James Perrett]

Right, raw files time:
Firstly completely unprocessed vocals from Gridlines (bit of clean gain to aid comparison): https://soundcloud.com/user-359526178/g ... al_sharing
Not crazily essey, but definitely needs work.

I only listened to the first few seconds before realising that the reason it needs work is because you are using totally the wrong mic for your voice. I'll bet it sounded wonderful to you in the shop but mics that sound wonderful in the shop don't necessarily work in real life.

An SM58 would work better than the mic you are currently using. In fact, it may even be the best mic for the job.

[Drew Stephenson]

Interesting. I don't actually own an SM58 but I'll try with an AKG D5 (closest I've got I think) and I may as well stick up my Beta 87 at the same time.
Thank you.

[Stuart79]

Right, raw files time:
Firstly completely unprocessed vocals from Gridlines (bit of clean gain to aid comparison): https://soundcloud.com/user-359526178/g ... al_sharing
Not crazily essey, but definitely needs work.

I only listened to the first few seconds before realising that the reason it needs work is because you are using totally the wrong mic for your voice. I'll bet it sounded wonderful to you in the shop but mics that sound wonderful in the shop don't necessarily work in real life.

An SM58 would work better than the mic you are currently using. In fact, it may even be the best mic for the job.

What makes you say this?

[SafeandSound Mastering]

When the esses are too much de-ess more, when they are lispy de-ess less.

Just do that and you cannot really go wrong.

If you can no longer judge go back the next day.

Unless the refs are essentially your own track and the same voice they won't really be very good references compared to using good judgement which the above is based on.

What mic is a new and different question. If voices are always ending up essy then yes capture the source better. You can soften a source a little using multiple foam pop shields and a couple of screen ones too.

That's for the future though, your vocal is already as is.

[Wonks]

D5 is quite bright compared to an SM58, so you may not improve things too much by using it. But it's worth a try.

[James Perrett]

I only listened to the first few seconds before realising that the reason it needs work is because you are using totally the wrong mic for your voice. I'll bet it sounded wonderful to you in the shop but mics that sound wonderful in the shop don't necessarily work in real life.

An SM58 would work better than the mic you are currently using. In fact, it may even be the best mic for the job.

What makes you say this?

I can hear a "zinginess" that I tend to associate with over bright large diaphragm capsules. It works on some, mainly female, voices but not so often on many male voices.

Drew - I've not used a D5 much but I think it is the mic that we used on the lead vocals for "I Hate This Party". (We were swapping vocal mics around all over the place so I'm not 100% sure)

[Drew Stephenson]

I'm hoping to have got rid of this cold by next weekend so I'll rig a selection up and see what comes out.

[Wonks]

I'm hoping to have got rid of this cold by next weekend so I'll rig a selection up and see what comes out.

Snot?

[Bob Bickerton]

So what mic did you use on your vocals Drew?

Bob

[Wonks]

I know he’s got a Warm Audio WA-87 R2, so I suspect it’s that.

[Drew Stephenson]

Hi Bob,
My main vocal mic for the last several years* has been a SubZero v12c: https://www.soundonsound.com/reviews/subzero-sz-v12c

I do have a WA87mkII but that really doesn't work with my voice at all; thin, reedy, spitty and generally unpleasant. The only other LDC I have is an sE2200 (first version) that's also on the bright side.

I'll make some time at the weekend to do a bit of a shootout of a few options and stick some samples up. Probably not much interest to most of our regulars but maybe for those with more limited budgets.

* I didn't use it for a few months when it was causing me problems but that turned to be because the power supply was sitting too close to another it of kit and picking up all kinds of nasties. Moved power supply, solved problem.

[Bob Bickerton]

OK, the closest frequency plot I could find is this: https://www.gear4music.com/Recording-an ... phone/154R (which I suspect is the same model). But really the D5 and Beta87 both have hefty HF lift as well and I find both to be 'edgy' rather than 'open'.

Anyway listening to the unprocessed track, it does sound the way James described, almost a ghost frequency/ies or sheen, which may well simply be the nature of that microphone, an attribute which is more common in condenser microphones.

I came across it with the Neumann TLM193, which I considered to be a neutral microphone (some would say dark), but when I put it up against a high quality ribbon (AEA N8) I could hear - very subtly - that condenser veil/sheen - actually I don't know how to describe it! They're still wonderful mics, by the way!

Bob

[tea for two]

Possibly a transformerless might do the trick.
I was considering CAD M179 glowingly reviewed by Hugh over 2decades earlier. It is transformerless also multipattern : two more reasons I was considering.
https://www.soundonsound.com/reviews/cad-m177-m179
It's available in thomann with their 30day returns policy wont break the bank.

[SafeandSound Mastering]

What De-esser are you using ?

[Sam Inglis]

Drew, if you ever fancy popping over to East Yorkshire I may have one or two mics you could try out...

[Matt Houghton]

It appears I'm a bit late to this thread....

1. You asked about references. I recall Mike Senior mentioning in an article many moons ago that he used Natalie Imbruglia's 'Torn' as a reference for de-essing —it's just a bit too sibilant so a good marker (if it's that bright, tame it a bit, if it isn't you're probably in safe territory).

2. You might find some useful technique tips in an article I wrote in SOS Sep 2020. In particular, the points about what you drink, blocking gaps in teeth and mic position.

3. It seems not all de-essers are created equal. I have to put a lot more work into setting some up than others. Why that is, I don't know.

4. I'd echo what people have said about mics and processing. Also, not sure anyone's pointed out that ribbon mics naturally have resonances much lower than typical capacitor mics, putting them out of the wanted audible range. It's why they seem to take EQ better, I think. So if you're doing a lot of stuff with high shelf boosts, exciters, saturation, compression etc, maybe that's something not think about. Not that all capacitors are bad by any means — there are many wonderful one. But controlling such things is probably a reason a lot of the good ones are pricey, and if it's an issue for you, maybe worth thinking about.

5. I'd bite Sam's hand off for that offer of auditioning mics if I were you...