Guess WHAT?! Yep Comp AGAIN! (before Desser or after?

[jaynm26]

I seen some mostly go after I have also seen some after corrective EQ (Ken Lewis, Tal) So what do you guys think the best place for it and why?

 

[mindsound]

Unless it is for a very sibilant vocal track (that may trigger too much the comp.) I'll put the comp first 'cause it will tame a bit the high freq. anyway. Then, it means that you may use the deesser subtly.

 

[Farview]

It all depends. A lot of times the compressor or EQ makes the ess's worse, so putting the deesser first makes sense. But after you do all the processing, if the ess's are still bad, put it at the end.

There is really no right answer, it all depends on what is going on with the specific track.

 

[miroslav]

These days I never even bother with a de-esser...I have a hardware unit that's been sitting in my racks unused for a long time, and AFA software plug-ins, I don't bother with them either, because once the vocal is in the DAW....I just go through the track and slice-n-dice wherever there is sibalance...then I manually adjust it either by lowering the volume of the "SSSS" portion or by applying some narrow bandwidth EQ cut in the appropriate frequency range...and sometimes I'll use both, because if you cut narrow frequencies to much, you can get weird sounding vocals, so a little narrow EQ cut and a little overall volume cut of the "SSS" and I can usually make most any sibilance sound perfect.

The other trick that is even easier at times is to simply go find a word with similar "SSS" sound...and just cut out the problems and splice in the good version.
I'll pull 2-3 really nice "SSS" sounds out of the vocal track...and just use them as many times as needed.
You couldn't tell the difference with a microscope or hearing aid....and in the end it sounds great.

With the editing power of a DAW....I've gotten more and more away from applying global processing on a track in order to fix specific issues. It's much more surgical to slice-n-dice and just zoom in on the trouble spots.
It takes a bit more time than just strapping a comp/EQ across the whole track...but IMO, much better results.

 

[Farview]

Wavelab has a spectral editor that lets you 'see' the problems and erase them with photoshop-type tools. That's another way to do it.

 

[Massive Master]

I *almost* always use the de-esser first in the chain if possible. It's basically being used as a fast-acting EQ. A compressor will "see" and use that excessive energy.

 

[jaynm26]

I *almost* always use the de-esser first in the chain if possible. It's basically being used as a fast-acting EQ. A compressor will "see" and use that excessive energy.

Explain how the compressor will use the excessive energy. (jus wanna make sure Im on right track.) So you feel having it first in the chain is better?

(I love hearing how people do stuff in such different ways and when you pick their brain you understand their logic behind its kinda like ....Ooohh!!! )

 

[Mo Facta]

Actually I find that before the de-esser helps the de-esser react more consinstently as the esses will be accentuated by the compressor.

Cheers

 

[Mr Clean]

These days I never even bother with a de-esser...I have a hardware unit that's been sitting in my racks unused for a long time, and AFA software plug-ins, I don't bother with them either, because once the vocal is in the DAW....I just go through the track and slice-n-dice wherever there is sibalance...then I manually adjust it either by lowering the volume of the "SSSS" portion or by applying some narrow bandwidth EQ cut in the appropriate frequency range...and sometimes I'll use both, because if you cut narrow frequencies to much, you can get weird sounding vocals, so a little narrow EQ cut and a little overall volume cut of the "SSS" and I can usually make most any sibilance sound perfect.

The other trick that is even easier at times is to simply go find a word with similar "SSS" sound...and just cut out the problems and splice in the good version.
I'll pull 2-3 really nice "SSS" sounds out of the vocal track...and just use them as many times as needed.
You couldn't tell the difference with a microscope or hearing aid....and in the end it sounds great.

With the editing power of a DAW....I've gotten more and more away from applying global processing on a track in order to fix specific issues. It's much more surgical to slice-n-dice and just zoom in on the trouble spots.
It takes a bit more time than just strapping a comp/EQ across the whole track...but IMO, much better results.

+1 to all of that. I've never used a de-esser.

 

[mixsit]

I've not needed dess (sometimes other similar chores with a multiband) but I'm curious too.
In the other thread we were kicking around how much (if..) the comp' will be reacting to 'level vs tone shape.
It would seem to me that if the voc isn't already in the ballpark level wise the dess' would be hit and miss.
But like I said- don't know, curious how folks see it.

 

[Farview]

The compressor doesn't really see the S's, it just turns down everything but the S's. So it makes them seem louder and worse. Knocking the S's back before compression will help. if you put it after the compression, it sometimes makes it harder to fix because the compression will make the S's longer and more pronounced.

 

[jaynm26]

The compressor doesn't really see the S's, it just turns down everything but the S's. So it makes them seem louder and worse. Knocking the S's back before compression will help. if you put it after the compression, it sometimes makes it harder to fix because the compression will make the S's longer and more pronounced.

Actually I find that before the de-esser helps the de-esser react more consinstently as the esses will be accentuated by the compressor.

Cheers

Fairview Mo says Comp>Desser helps the desser to react more consistently as the esses "WILL" be attenuated by the comp. We got 2 diff views here WHO IS RIGHT!!??? ...Im confused some more posters please weigh in please.

 

[Massive Master]

The easy part is to try it both ways. I tend to use the DESR (I think I'm going to start typing it like that) before anything if the sibilance is pretty blatant. But as mentioned, sometimes compression can make not-so-blatant sibilance stand out more, making it easier to tackle post-compression.

 

[ashcat_lt]

I think sometimes we're over sensitive to what we consider sibilance. S (and F) are natural parts of our language, and don't necessarily need to be obliterated every time they pop up. It's only a problem if it's a problem, and I'm not sure that it's a problem as often as some folks think. But sometimes it is...

Of course the best way to deal with sibilance is in tracking. The vocalist should learn to de-emphasize those sounds one way or another. It takes conscious effort to avoid "hanging" on an S sound. There are ways to shorten or soften those sounds. There are also little tricks like a slight turn of the head. If you cant get the vocalist to do these things you can sometimes try a pop filter or try angling the mic some.

But that's not the question. My ex had so much metal in her face, and kinda funny dentition that made her S sounds into rapiers! Asking her to change her "technique" in any way generally turned into telling her she completely sucks. So, I had to work with it! For her I usually tried to use "slower" dynamic mics and then would automate the really bad artifacts as mentioned above.

I guess I can see where having the overall track more consistent could allow the de-esser to act more consistently. OTOH - the S sounds really are usually pretty quiet compared to the rest of the vocal. There's maybe a bunch of high end but very little meat, so it shouldn't really be touched by the compressor. They will be louder in comparison after though... I guess the only right answer is to do what works best for each track. If it don't work one way try it the other.

And now that I've made a semblance of a contribution to the thread... I really try not to publicly correct spelling and grammar - proper nettiquette says to let it go since we literally have everybody in the world, some who don't speak fluent English, some typing real fast on mobile devices, whatever. This one really bugs me though! The way we usually use it, the phrase "guess what" is an imperative statement - telling someone to do something - and not an interrogative statement - asking a question. The ? is always inappropriate. An ! Is much better. The OP here used both, but the ? Comes first.

If I was to say "Guess!", and you reply "Guess what?", then the ? works. That almost never happens, though.

Sorry to be "that guy". Please take it as constructive criticism. Don't be like the ex!

 

[bouldersoundguy]

De-essing should be done before the sound hits the mic.

 

[miroslav]

De-essing should be done before the sound hits the mic.

Yeah, but that's not always realistic or easy.

You can put up a pop filter, tape a pencil to the microphone, have the singer stand off-axis, and any other trick...and sometimes the "SSSSS" sounds just end up being too pesky when you drop the vocals into the mix even though they are not very bad when recording.

 

[bouldersoundguy]

Yeah, but that's not always realistic or easy.

You can put up a pop filter, tape a pencil to the microphone, have the singer stand off-axis, and any other trick...and sometimes the "SSSSS" sounds just end up being too pesky when you drop the vocals into the mix even though they are not very bad when recording.

True. If it's not sitting in the mix and I felt the need to process I would probably try editing first. A de-esser might go at the front or at the end of the chain depending on what else was happening. A multi-band compressor can also sometimes do the trick.

But if someone needs de-essing it's usually because of something different about the way they sing. I've found that people with a theatrical background tend to over-enunciate their consonants when using microphones. This leads to all sorts of trouble with sibilance and plosives. But theater people also know how to take direction, so I ask that they not push their consonants so hard. Good monitoring really helps. Singers generally need a fair bit of volume so they hear the mic over the bone conduction, otherwise they will not hear their own sibilance as it sounds to the mic.

 

[Farview]

Fairview Mo says Comp>Desser helps the desser to react more consistently as the esses "WILL" be attenuated by the comp. We got 2 diff views here WHO IS RIGHT!!??? ...Im confused some more posters please weigh in please.

I've posted twice in this thread with arguments for both ways. It is completely situational. It depends solely on the signal, the other processing and how it sits in the mix.

 

[Beck]

Since a de-esser is a compressor (frequency dependent), sometimes that's all you need, but if using both (and I'm talking hardware in particular here) I generally use the de-esser first, primarily on vocals and drum overheads. However, if its late in the project and you find you've wound up with a little more sibilance than expected then of course you can apply the de-esser at that point.

 

[Mo Facta]

In that case, EVERYTHING is situational and questions like this are unaswerable.

What are we all doing here????

Cheers