De-essing

  • Thread starter Thread starter tvolhein
  • Start date Start date
T

tvolhein

Tom Volhein
Hi,

I have a sibilance problem on a vocal track. I have tried Spit Fish with no luck. I think that it is because I am running 64-bit and Spit Fish is 32-bit.

I need either another de-esser or process to de-ess the problem myself.

Any suggestions? I can't re-record the vocal track.

Thanks in advance.

Tom
 
As far as i know a de esser is just a compressor that focusses on specified higher frequencies, so if you have a multiband compressor you could use it to the same effect.

I kinda wanna experiment with duplicating a track, sticking a high pass on there, a noise gate and inverting the polarity.

Pretty sure it would suck though,lol. Just thinking out loud.
 
Volume automation (Pre FX)

sibliance is pretty obvious on a waveform, just draw in automation around the Esses to drop the volume at that point. Make it pre FX so it doesn't mess up other comps, eqs, reverbs, etc

sounds far more natural than de-essing and doesn't leave nasty artifacts or effect the whole vocal

If I can't retrack that's how I always go rather than use a de-esser, takes time but it's worth it

Another approach would be to cut the esses to a separate track and then use a high Q surgical EQ to drop the offending frequencies but I still prefer volume automation
 
Steenamaroo said:
As far as i know a de esser is just a compressor that focusses on specified higher frequencies, so if you have a multiband compressor you could use it to the same effect.

I kinda wanna experiment with duplicating a track, sticking a high pass on there, a noise gate and inverting the polarity.

Pretty sure it would suck though,lol. Just thinking out loud.
Click to expand...

You have a dizzying intellect :D
 
Bristol Posse said:
Volume automation (Pre FX)

sibliance is pretty obvious on a waveform, just draw in automation around the Esses to drop the volume at that point. Make it pre FX so it doesn't mess up other comps, eqs, reverbs, etc

sounds far more natural than de-essing and doesn't leave nasty artifacts or effect the whole vocal

If I can't retrack that's how I always go rather than use a de-esser, takes time but it's worth it

Another approach would be to cut the esses to a separate track and then use a high Q surgical EQ to drop the offending frequencies but I still prefer volume automation
Click to expand...

+1

I haven't found a de-esser plug-in that doesn't mess with the sound of the track. Volume automation is the only process I've used that dosen't effect the sound of the track.
 
Steenamaroo said:
As far as i know a de esser is just a compressor that focusses on specified higher frequencies, so if you have a multiband compressor you could use it to the same effect.
Click to expand...

That's what I've done (used Reaper's built in Recomp).
I've found that you have to tune it to each person's voice though.
 
Same here.
i
I ve used TONS of de-essers. None of them really satisifed me. They always make the track sound dull.
I automated the Ss by hand now. I takes time but the result is garanteed.
 
Surely there's a market for this.
I mean, waves now have vocal rider and bass rider. Shouldn't they have an Srider?

Also, isn't that someones username on here?
 
An old trick is to not pronouns the S but make a Ch sound instead!
 
I just read about this a few days ago, and it works really well.

1) copy vocal track so you have an original and a copy
2) find frequency where s is jumping out on the copy track (usually around 5k)
3) boost the EQ on the copy track by 6db or more (it will sound horrible)
4) put compressor on original track
5) feed compressor with copied track
6) adjust compressor to taste
7) make sure you muted the copied track

The compressor on the original track will only kick in when the s jumps out. so your vocal track will be unmolested when there's not an s.

make sense? I just started doing this, so my explanation may need some refining. basically this is making your own multiband compressor, but you have tons of control over it and it will work well on any voice with some adjustment.
 
I tend to use a de-esser and also volume automation to get rid of sibilance. Sometimes volume automation is enough on it's own and sometimes a de-esser is enough, but on tracks with loads of sibilance I tend to use both techniques.

G
 
I just don't sing songs with esses in them:

like Tareway to Heaven and ...
 
I read in that Roey Izhaki mixing book about how compressors can act as de-essers and listening back to a vocal track my friend did about 11 years ago that I recorded, I can tell I'd just gotten the compressor and that I had no idea how to use it {not much has changed there ! } because there's places where her 's' has completely disappeared ! I have to laugh. It sounds so weird hearing someone sing the word 'mysterious' with no Ss in there !
 
grimtraveller said:
I read in that Roey Izhaki mixing book about how compressors can act as de-essers and listening back to a vocal track my friend did about 11 years ago that I recorded, I can tell I'd just gotten the compressor and that I had no idea how to use it {not much has changed there ! } because there's places where her 's' has completely disappeared ! I have to laugh. It sounds so weird hearing someone sing the word 'mysterious' with no Ss in there !
Click to expand...
Ah the good old "turn the singer into someone with a lisp" setting. :)

G
 
I have tried many many many de-essers in the past. I never was 100% satisfied with the result. I always found that is was more productive to de-esse by hand instead of tweaking a plugin forever and ever...

if you still want to use a plug... How to use a De-esser on vocals - YouTube
 
aaronmcoleman said:
I just read about this a few days ago, and it works really well.

1) copy vocal track so you have an original and a copy
2) find frequency where s is jumping out on the copy track (usually around 5k)
3) boost the EQ on the copy track by 6db or more (it will sound horrible)
4) put compressor on original track
5) feed compressor with copied track
6) adjust compressor to taste
7) make sure you muted the copied track

The compressor on the original track will only kick in when the s jumps out. so your vocal track will be unmolested when there's not an s.

make sense? I just started doing this, so my explanation may need some refining. basically this is making your own multiband compressor, but you have tons of control over it and it will work well on any voice with some adjustment.
Click to expand...

so you just sidechain the boosted, muted track to a compressor over the unmuted main vocal track? sounds like it would work, might try it thanks
 
You must log in or register to reply here.
Back
Top