Can't get rid of 'S' noise in my vocals
- Thread starter Crysiq
- Start date
Massive Master
www.massivemastering.com
Or in the "singing technique" forum. 
Armistice
Son of Yoda
Yes... don't try it in the mastering stage... it's a tracking / mixing issue.
If you can't re-record a better take, put the de-esser away if it's not working, solo your vocal track, expand the wave form vertically and horizontally until it's approximately huge, and automate a volume cut (a simple V works most times for me, but if it's a long S sound you might need to have a reverse plateau shape) of 5 db over the sibiliance..., then unsolo your vocal track and listen - see what you think, and adjust if you need to.
You'll get pretty good at working out the exact length and depth of your cut - I can pretty much do them all in a single play through as long as it's not a particularly essy set of lyrics.
Works better for me than any de-esser I've found in the "free" range... YMMV...
No magic bullets, sorry, we're out of stock. Work required instead...
If you can't re-record a better take, put the de-esser away if it's not working, solo your vocal track, expand the wave form vertically and horizontally until it's approximately huge, and automate a volume cut (a simple V works most times for me, but if it's a long S sound you might need to have a reverse plateau shape) of 5 db over the sibiliance..., then unsolo your vocal track and listen - see what you think, and adjust if you need to.
You'll get pretty good at working out the exact length and depth of your cut - I can pretty much do them all in a single play through as long as it's not a particularly essy set of lyrics.
Works better for me than any de-esser I've found in the "free" range... YMMV...
No magic bullets, sorry, we're out of stock. Work required instead...
Steenamaroo
...
+1 to that! ^^
jimmys69
MOODerator
Moved thread...
I would first address why the 'esses' are an issue. A pop filter (or two) is one thing that is good to help with that. Also addressing the mic at an off angle to the diaphragm is another. The most important one that is most obvious to change is the performance itself.
Fixing things like this after the fact, is even harder to do. You either have to spend a bunch of time editing them out, or spend money on a plugin that may-or may not do it for you.
Best to get the performance sweet before using any plugin to 'fix'.
I would first address why the 'esses' are an issue. A pop filter (or two) is one thing that is good to help with that. Also addressing the mic at an off angle to the diaphragm is another. The most important one that is most obvious to change is the performance itself.
Fixing things like this after the fact, is even harder to do. You either have to spend a bunch of time editing them out, or spend money on a plugin that may-or may not do it for you.
Best to get the performance sweet before using any plugin to 'fix'.
miroslav
Cosmic Cowboy
Yes...DITTO ^^^^
Soimetimes I'll use some narrow EQ cut in the frequency area of the "ssss" sound....like if there's whistling with the "sssss", but it's not much of a loudness issue....or a combination of both.
Not sure how friendly Audition is though for slice-n-dice type of editing....???
bouldersoundguy
Well-known member
Or in the "singing technique" forum.
This. Singing it right is the best solution, with fixing it in the mix a distant second.
ashcat_lt
Well-known member
The first step is to get some perspective and make sure that it's actually an issue. I think a lot of people have the instinct to go a bit overboard with anti-sibilance treatment, and the OP saying "get rid of..." really worries me. We tend to get all blindered and nitpicky sometimes, and on top of that a lot of the lower end speakers can be a bit hyped or harsh right in that sss area, and this can be exacerbated by problems in the room itself.
My monitors are notoriously unforgiving in the region that makes sibilance really hurt, and I struggled on my last album up to the mastering stage. Eventually I convinced myself that some really great sounding mixes that I know quite well were coming across a bit harsh also. So, like, try that real quick. Listen to some other stuff in the same general genre that you really like on your system, and try to be as critical of their mix as your own. Or try listening on a number of other systems just to make sure that it's not just your speakers/room.
My monitors are notoriously unforgiving in the region that makes sibilance really hurt, and I struggled on my last album up to the mastering stage. Eventually I convinced myself that some really great sounding mixes that I know quite well were coming across a bit harsh also. So, like, try that real quick. Listen to some other stuff in the same general genre that you really like on your system, and try to be as critical of their mix as your own. Or try listening on a number of other systems just to make sure that it's not just your speakers/room.
Armistice
Son of Yoda
Good point... I've noticed several recordings lately, professionally done of people whose music I dig where the sibilants are probably more obvious than I have on my stuff via reducing it using the method I mention above. You need to watch the reverb as well... can make it worse.. if you're drowning the lead in reverb, it's going to be really obvious...
BroKen_H
Re-member
I usually use a sidechained compresser, and just skip the deesser altogether. But that's my voice...
This is a Propellerhead tutorial, but it has a lot of broad, useful information on vocal recording.
Here is a tutorial for Reason (what I use), but you can probably find a video tutorial on sidechaining to de-ess on YouTube for your DAW.
This is a Propellerhead tutorial, but it has a lot of broad, useful information on vocal recording.
Here is a tutorial for Reason (what I use), but you can probably find a video tutorial on sidechaining to de-ess on YouTube for your DAW.
dainbramage
New member
Yes... don't try it in the mastering stage... it's a tracking / mixing issue.
If you can't re-record a better take, put the de-esser away if it's not working, solo your vocal track, expand the wave form vertically and horizontally until it's approximately huge, and automate a volume cut (a simple V works most times for me, but if it's a long S sound you might need to have a reverse plateau shape) of 5 db over the sibiliance..., then unsolo your vocal track and listen - see what you think, and adjust if you need to.
You'll get pretty good at working out the exact length and depth of your cut - I can pretty much do them all in a single play through as long as it's not a particularly essy set of lyrics.
Works better for me than any de-esser I've found in the "free" range... YMMV...
No magic bullets, sorry, we're out of stock. Work required instead...
Cool, I never thought of that. I have a track that seems pretty hopeless S-wise. The vocalist is on the other side of the country for long, so I'm gonna try that too.
Farview
Well-known member
My advice for the vocalist is to sing louder. That is usually the problem. Consonant sounds tend to be about the same volume all the time, so if the vowel sounds are loud, the consonants sound quieter in comparison. If the singing is really quiet, the s,t, p and b sounds will seem loud and out of place.
TheChikenMaster
I'm Unbalanced
I would say move the mic back(if you can). Air makes the best compressor. And it will give more room for the vocal to breath naturally. giving you that 'transparent' sound.
Greykitkat36
New member
I've had a lot of success in this area using EQ rather than the other plugins or techniques mentioned. Obviously if you can re-do the take at the recording stage, that is without a doubt the best way to correct it. However, if you can't re-record or have time constraints etc. here's my suggestion.
Rather than using a De-esser which can really sound unnatural at times depending on the settings (sometimes you have to push it pretty hard if the sibliance is very bad) you can try and pinpoint it using EQ sweep. Lower your Q to pretty narrow--less than 1 and search for the area where it's very bad or stands out a lot. For males this tends to be in the 3k-8k range. Once you find that spot, cut that frequency by 1-3db and expand the Q a tad for a more natural cut, and see if that helps at all. I've noticed this can help tame those harsh "Ss" while at the same time keeping your vocal bright and clear.
Rather than using a De-esser which can really sound unnatural at times depending on the settings (sometimes you have to push it pretty hard if the sibliance is very bad) you can try and pinpoint it using EQ sweep. Lower your Q to pretty narrow--less than 1 and search for the area where it's very bad or stands out a lot. For males this tends to be in the 3k-8k range. Once you find that spot, cut that frequency by 1-3db and expand the Q a tad for a more natural cut, and see if that helps at all. I've noticed this can help tame those harsh "Ss" while at the same time keeping your vocal bright and clear.
RecordingMaster
A Sarcastic Statement
I remember Andrew Schepps saying he doesn't de-ess much if at all. He has his ASSistants do volume automation rides (probably drawn in pro tools rather than manually ridden) on S sounds.
Just a more natural alternative to de-essing - less obtrusive to the rest of the signal and might work better for the task at hand! I do this for the double on a lead vocal if they are singing the exact same thing (actually I just delete the first half of the doubled S and do a quick fade in from there - sounds unnatural by itself but you don't notice when the full S is there on the top lead) but on the top lead I will usually just lazily de-ess with a de-esser!
Just a more natural alternative to de-essing - less obtrusive to the rest of the signal and might work better for the task at hand! I do this for the double on a lead vocal if they are singing the exact same thing (actually I just delete the first half of the doubled S and do a quick fade in from there - sounds unnatural by itself but you don't notice when the full S is there on the top lead) but on the top lead I will usually just lazily de-ess with a de-esser!
PDP
There once was a note
I manually edit all the "s" sounds usually by -5db. It isn't that bad once you get good at it, it actually goes pretty fast. The other thing is learning to sing with less "s" pronunciation. I don't compress either, but I do use a lot of high end on my voice, so as kitkat said you can find the harshness between 3-8k and make a small cut. Mine is right at 4k.