Best De-esser

[Ploki]

I don't like melodyne's sibilance detection. I generally disable it.
If you pitch something and sibilance is attached to the pitch note, you need to split it, or else it will absolutely destroy said sibilant. not great

That was how it worked before Melodyne 5. After v5, Melodyne automatically isolates the sibilant from the rest of the "blob", so when you change the pitch or the note length, the sibilant is completely unaffected. The Melodyne 5 sibilance tool absolutely smokes any traditional de-esser I've tried.

I have 5 Studio. And no, it doesn't always detached the sybilant. + It sometimes detects airy passages as sibilants.
i find Pro-DS infinitely more consistent

[trackbout]

The only time I've ever needed to guide the detector is if I've selected an entire vocal track at once, in which case you may catch a non-sibilant that is also non-pitched information, which is largely what the detector is keying on. If you don't select the entire track at once and just work on the problem areas individually, which is usually what I'm doing since I rarely want to make track wide sibilance decisions anyway, that's not an issue.

Anyway detection stage aside, once a sibilant has been properly designated, the sound quality is always better than a de-esser to my ears, because you don't have to make the compromises that you have to make with any traditional de-esser, including FabFilter's, which is acting on a signal simply based on frequency and amplitude, so it will always have an impact on the overlapping, non-sibilant signal. Melodyne can act on a sibilant without touching the melodic content.

[Ploki]

The only time I've ever needed to guide the detector is if I've selected an entire vocal track at once, in which case you may catch a non-sibilant that is also non-pitched information, which is largely what the detector is keying on. If you don't select the entire track at once and just work on the problem areas individually, which is usually what I'm doing since I rarely want to make track wide sibilance decisions anyway, that's not an issue.

Anyway detection stage aside, once a sibilant has been properly designated, the sound quality is always better than a de-esser to my ears, because you don't have to make the compromises that you have to make with any traditional de-esser, including FabFilter's, which is acting on a signal simply based on frequency and amplitude, so it will always have an impact on the overlapping, non-sibilant signal. Melodyne can act on a sibilant without touching the melodic content.

Fab Single vocal algo detects sibilant nearly as accurately as melodyne.
Better even in my experience.
Airy/whispered vocals (think billie eilish) fall apart with melodynes sybilant algorithm because a lot of things are interpeted as breath / sybilants.

If you enable monitoring on fabfilter singlevocal and bounce in place you’ll be left with nothing but sybilants. Extremely robust.

Im not fond of melodyne’s variant.
Furthermore you usually need to de-ess post compression as well, and you generally don’t pitch compressed vocals.

[Adrian Charras Music]

Classic Waves De-esser

[Cooker]

Waves sibilance

[Winstontaneous]

I work with a male vocalist who has an incredibly sibilant but expressive voice. I’ve used some combination of Oeksound Soothe, ToneBoosters Sibalance and the Dynamic EQ in EQ4, FabFilter DS, Metric Halo De-Esser, TDR De-Edger and the smart de-resonator on Nova EQ GE, Voxengo Voxformer, and other tools I’ve surely forgotten. None of these took care of the issue all the time on its own.

If you have control over the recording process and esp. if the sibilance is in the 8k to 12k range, be aware that many large diaphragm condenser mic capsules (especially cheap ones) have a significant boost at these frequencies, so a dynamic vocal mic (like a Shure SM7B with the big foam windscreen) can be a better bet. And saturation/distortion or other processing that adds upper harmonics is best avoided, or at least applied later in the chain after de-essing is performed.

Also to my ears, manually automating gain on sibilant phrases is often the most transparent approach of all. There are also tricks like taping a pencil to the mic (more for p plosives) or having the vocalist face slightly away from the mic so the more directional high frequency noises are less likely to get picked up.

[Digivolt]

The Accusonus ERA stuff was/is pretty good

[jamcat]

The only time I've ever needed to guide the detector is if I've selected an entire vocal track at once, in which case you may catch a non-sibilant that is also non-pitched information, which is largely what the detector is keying on. If you don't select the entire track at once and just work on the problem areas individually, which is usually what I'm doing since I rarely want to make track wide sibilance decisions anyway, that's not an issue.

Anyway detection stage aside, once a sibilant has been properly designated, the sound quality is always better than a de-esser to my ears, because you don't have to make the compromises that you have to make with any traditional de-esser, including FabFilter's, which is acting on a signal simply based on frequency and amplitude, so it will always have an impact on the overlapping, non-sibilant signal. Melodyne can act on a sibilant without touching the melodic content.

Fab Single vocal algo detects sibilant nearly as accurately as melodyne.
Better even in my experience.
Airy/whispered vocals (think billie eilish) fall apart with melodynes sybilant algorithm because a lot of things are interpeted as breath / sybilants.

If you enable monitoring on fabfilter singlevocal and bounce in place you’ll be left with nothing but sybilants. Extremely robust.

Im not fond of melodyne’s variant.
Furthermore you usually need to de-ess post compression as well, and you generally don’t pitch compressed vocals.

I would separate the audio into 2 tracks: sibilance and tonal, and process them separately, whether I was using Melodyne or something else. This allows you to control the volume separately and apply compression, reverb, or pitch to only the body of the vocal.

[simon.a.billington]

Waves sibilance

This would be one of my picks too. Managed to grab it for free during one of their Black Friday promotions. It's very good, natural sounding.

My other pick would be the RX Desser in the Spectral Mode. It uses quote a bit of CPU and ads a fair bit of latency, but you could always freeze your tracks.

[kraster]

Oxford SuprEsser has been my go to for years.

UI is a bit scruffy nowadays but it still works well.

[audenso]

I’m mostly using DMG Audio Essence and my second deesser of choice is FF Pro-DS.

[Weasel-Boy]

Try as I might, I’ve never found a software de-esser that works as well as my old hardware DBX 263x. I still use it to this day, via an external send/return loop into my DAW. Basically, a one knob affair that just works. Yes, not a convenient plug-in solution…but an excellent sounding alternative for all kinds of troublesome sibilant vocals.

[Songwriter101]

Volume automation (clip gain) works best for me. Don't mind taking the time to do that. Other than that Esspresso by Klevgrand gives me great results.

[yzcoruhT]

This one:

https://www.google.com/search?q=frank+p ... e&ie=UTF-8#

[ShotaroMusic]

I’m mostly using DMG Audio Essence and my second deesser of choice is FF Pro-DS.

Same! I mainly use those two as well.