De-esser test.

It may have puzzled some of us that the vocal range of the music we were mixing would have been basically fine, but the s of the singer's lyrics were so prominent that it was seriously offensive to the ear. In a lucky case, a smooth, tight EQ can solve the problem, but in more extreme cases this solution is inadequate. What can be done in such cases? This is where so-called De-Esser applications were invented. But it makes a difference what software you use. The different types of software vary considerably in terms of their effectiveness and knowledge. In this article, we will look at some free and paid software.

Free:

Modern De-Esser

I'll be honest, it's weak. It works to some extent, but it also takes too much of the surrounding frequencies. It works like a compressor that only works in a given frequency range, but unfortunately it doesn't work in reality. It can be used in limited situations. Spitfish

Spitfish

A more controllable application. The "tune" value should be set higher, specifically to the highest frequency range, otherwise it cuts too much of the frequencies below it. This also means that if you set it to 8 kHz, it will not only cut around the 8 kHz value, but everything upwards. Still, with a little tweaking it can be used in most cases.

Tonmann

Although there are more buttons on the app, this gives you a bit more freedom in the settings. Of all the software so far, this is the one that handles the frequency cutting best.

Sleepy time DSP Lisp

I was pleased to see that the software handles mid/side mode. If you have a stereo music vocal with teng too many s,s,z's, there's not much you can do, but with the mid/side mode you can make it so that you only extract unwanted frequencies from the mid vocal, but not the edges. This program can do that. However, as crazy as it is, the intervention can't be intense enough at the most extreme value. In practice, this means that the software's creator must have thought that as much interference as the software allows is enough for a given energy band. In most cases, it is enough, so this software will achieve its goal, but if there is too much confusion, it may not produce a perfect result.

In the free section, Tonmann and Lisp are the favourites. The advantage in favour of Lisp is the use of M/S Left/Right mode, otherwise both programs are about equally effective.

Paid.

Waves De-Esser

Works well, does the job: maybe a bit softly into the track, i.e. doesn't allow more aggressive cutting. Most of the time it's not needed. If you like simplicity, well you won't regret using it. That's one advantage over its free counterparts, but for more complex interventions it's not enough. In any case, it is more efficient than the next software:

Izotope Nectar

It is not a de-esser application only, but a complex package that allows you to freely manipulate the raw izotope up to its finished state. So this is a really good software with a lot of features and De-Esser features. Basically, it can handle most tasks, but there is no mid-side, left-right mode and if you check the cut spectrum with the "ess-only" button, you can hear that the cut is too much into the lower frequency range even at the highest cut. Also, here we can only enter the lower limit on the frequency filter, not the upper limit. The Nectar service is therefore only basic in this respect. I have tried to use it many times, but I have had to switch several times because I am disturbed that I cannot get what I want to hear out of it.

Fabfilter Pro DS

I'll end with the software that is the best of all the software listed on the market. The frequency range you want to cut can be defined from "from to". For stereo music, you can set it to cut only the high frequencies in the middle or at the edges. If you want, you can completely cut out all hissing notes from the vocals. At normal values, the cut gives a soft enough feel to keep the treated track at a natural level. If you are spending money on De-Esser, I recommend this.

What can you do if you don't have De-Esser to hand? Obviously in simple cases we can download a free version from somewhere, but if for some reason we don't want to do that, we can also achieve results by using a multiband compressor. The key is to be able to reach into narrow spectrums of each frequency band. If it doesn't do that, then unfortunately you're not just cutting out the hissing sound from the track.

Using the Isotope Ozone dynamics compressor, I set up a narrow spectrum and compressed it to the extreme. With this solution, an average result can be achieved, which should be enough in most cases. Another option is an Auto Eq, where you set the desired frequency range, but the volume reduction is not proportional to the EQ setting, because the volume reduction is done by an algorithm similar to the capabilities of a compressor. The downside of this is that you can only hear what you've done, but you can't downsample that range to hear what you've actually cut out of the audio material.