ApulSoft splitS a more effective way to de-essing

[Frankie.T]

In my opinion this is the best release of the year.

I work with lots of music that comes with vocals, and de-essers never got it right, and i end up to manually edit every single esses and tees, and it's a tedious work to do!

Ofc it always depends on the take you are working with, different singers, preamps, microphones, and whatever happens in the chain and the recording enviroment.

But if happens to work with some home made vocal, it's usually a mess.

When i saw this plugin was like "is this real", let me try it!

And i was surprised, because it works very well.

Anyway this is the product page, where you can take a look and try the demo aswell
https://apulsoft.ch/splits/

Regular price: $59
Intro price: $49

If you want to watch the plugin in action and to know a bit more, i prepared a short video you may hopefully find useful

[hightyde]

I mean it really doesn’t seem all that different than antares deesser, am I wrong? Either way I’m sure if it does the same thing. Then it’s already a step better than most deessers.

[Frankie.T]

Never tried Antares, i don't like their business model, neither the way they behave with clients.
Try to get in touch with them, or make a question in forums when they come to write about a new release.
It's easier to have a talk with the pope.

[hightyde]

That’s completely justified, Frankie. Looking at the two. It seems they operate very similar.

[Dr.Gunjah]

I wonder why they stopped there and didn't add some logic to treat louder sibilances stronger

[Frankie.T]

I wonder why they stopped there and didn't add some logic to treat louder sibilances stronger

I think this is very difficult to implement, if not impossible for an online processing.

To do a good job like you suggest, you'd need to analyze the whole track, so you can detect and analyze any S, and so let the program to calculate S lenght, loudness (rms and/or peak?) and so to make them at the same loudness level, than you slide at the level you want.

As i suggest in my video, probably the best things is to place this de-esser after compression, so you don't just even the "good" part of the vocals, but also the S and so if their volume is even, you can easilly have one setting that works for everything.

Anyway maybe in the future with some AI that decide in advance the perfect loudness level of S and it adapt every incoming one to match that, could be what you are looking for, and what would certainly be an interesting improvement.
In that case i would just want to have the last decision about the S level though

[Dr.Gunjah]

What i meant was basically compression, but only for the S. So a threshold and ratio which adjusts the (negative) gain you set in relation to how much the S exceeds the threshold. There already is the waveform display it should be easy to let the user move a threshold line in that display for easy adjustment.

[yzcoruhT]

Anyone using it? Wanna demo it...

[trmupstage]

I was just looking at this as well. Do I need it? No, but I'm still curious as to what it can do for me to speed up handling sibilance

[fisherKing]

i use several different plugins to deal with sibilance, and splitS is one of them; it works well (altho yes, would be nice if it were dynamic).