Eiosis E2Deesser by Funkybot
Formats: OSX / Win, 32/64bits, AAX, VST2, VST3, AU
Let’s be honest, we don’t use de-essers because we want to. We use de-essers because we need to. By nature, they’re designed to be problem solvers. In a world full of bright microphones, singers with less than stellar microphone technique, sub-par acoustic environments, heavy compression, saturation, and liberal high-end EQ boosts, there’s a lot of opportunity for siblant sounds in vocals to get harsh. And that’s just vocals! The same can be true for other potentially bright mix elements like cymbals or squeaky guitars. You may even get handed a 2-channel mixdown that has some hard ess sounds and/or cymbals where going in and remixing isn’t always an option.
It’s not always practical or even desired to fix the root cause of those problems individually, so many years ago, some bright soul realized you could cut the s sounds out of a vocal, and start treating them independently to help reduce the harshness. At some point, dynamic EQ’ing became a thing and hardware de-essers starting hitting the market to try to automate the process.
Eiosis, the company founded by Fabrice Gabriel, of Air EQ and Slate Digital fame, recently released their e²deesser with the goal of being the ultimate de-esser. Let’s see how well they succeeded.
Features & UI
The e²deesser was designed to be as simple, or complex as you want it to be. If all you want is some quick vocal de-essing and can’t be bothered to get into the more advanced features, then set your Mode to Solo Voice and leave everything at the default settings, then adjust the Sensitivity and Amount. On the other end of the spectrum, if you’re the kind of person that wants to process your s sounds completely independently of the “voiced” signal (as it’s referred to here), e²deesser can also be the de-esser plugin for you.
As I mentioned, the Mode, Sensitivity and Amount controls are all you need for basic de-essing. The Mode dropdown includes options for: single voice, background vocals, voiceover, overheads, guitar squeaks, along with stereo, m/s, and mid mastering modes. Sensitivity dictates how acute the response to siblants is, while the Amount knob dictates the amount of reduction.
Below the big knobs are some additional controls for some extra tweaking. The Auto knob will dynamically adjust the frequency response, looking for any peaky frequencies and the Smooth knob introduces some saturation to the siblants to help tame them. Both are interesting features and can really help fine-tune the de-essing. The Output section has knobs for Gain and a simple Dry/Wet mix control.
The lower righthand section of the UI is where things get really interesting. This section will cover your 1) wideband de-esser, 2) split-band de-essing (dynamic EQ), 3) voiced signal EQ’ing, and 4) monitoring controls. Like I said, if you’re into the idea of having complete control, this is the plugin for you.
The red horizontal line that runs across this section represents your wideband for classic wideband de-essing. The amount of reduction that occurs isn’t just determined by the Amount knob, but tweaking the Auto knob will actually change what sounds get pulled into this reduction. Once you start adding some EQ bands to the red path, you’re essentially moving into the split-band territory. This is fantastic because it allows you to basically dynamically EQ the ess sounds, completely independently of the voiced signal. Notching in on the ess frequency is a great way to improve the quality of the de-essing, and on some vocals, low-passing the esses can be just what the doctor ordered.
The blue path represents your voiced EQ. Like I mentioned, e²deesser allows you to treat your esses and voiced paths completely independently and basically includes a built in EQ with the Eiosis Air band, and options for bell and shelf EQ’s. I’ll be honest, if this had a HP filter, you could get away with using another EQ entirely.
The monitoring section includes some handy tools to listen to the voiced signal, just the siblants, or a “siblant focus” mode that mixes in some voiced signal for context, but reduces the volume. If you wanted to completely isolate your esses from your vocals, you could always duplicate the track in your DAW and use the different monitoring modes to get one track with just the voiced signal, and another with just your esses. Very cool.
The top of the UI features a handy (and collapsible if you’d rather use your ears) time-frequency display that shows your frequencies on the X axis, and uses darker/lighter shades of blue to indicate intensity. Once the Sensitivity starts picking up transients, yellow bands will start appearing letting you know exactly what the sensitivity knob is responding to. Any EQ points you add will appear as horizontal lines across the frequency axis, which lets you know if you’re notching or boosting in the right places.
The menu includes the ability to adjust the lookahead, including a zero-latency option. There’s also the option to save a new default preset, which was a nice touch, and will come in handy for people using hosts that don’t allow you to overwrite the default preset. You’ve also got some spectrograph timing options up in the menu.
The first thing I noticed when testing the e²deesser out on some vocals was that the default 12 o’clock settings for the Sensitivity and Amount knobs seemed a bit heavy-handed for my tastes. Maybe it’s just me, but like adding salt to taste, I don’t want to put a de-esser on a track and hear it immediately responding to the vocal. I’d rather turn up the Sensitivity and Amount as needed while the track plays. This is where the ability to change the default preset was very appreciated.
For the sake of my tests, e²deesser was being pitted up against the Fabfilter Pro-DS, which has been my go-to vocal de-esser for a few years now. The Pro-DS has long been a favorite of mine because it “just works.” I throw it on a vocal, leave it on Single Vocal mode, wideband, adjust the amount and threshold and I’m done. The question for me here was, can the e²deesser work as easily?
The first vocal I put up was for a very high-tempo (248 bpm) punk tune with a male vocal. This is a bright mix, with a bright vocal and a singer whose register is in the Ray Davies-range. There were quite a few hard esses, and e²deesser sounded absolutely fantastic in this vocal. Starting out with the wideband produced very similar results to Pro-DS, but the ability to notch and lowpass the esses really gave e²deesser a slight edge over the two.
One thing I noticed about the e²deesser is that I found myself tweaking the various settings a lot more than Pro-DS. The temptation to “make it sound even better” kicks in due to all the options, so I’d find myself not stopping at “good enough” and really digging into the controls to see how they’d interact with the vocal. Part of this was absolutely because e²deesser is new and I’m still learning it, but sometimes having lots of bells can get in the way, which I think factored in here. In honesty, the Pro-DS vocal was good enough, but the extra tweaking done on the e²deesser did result in a slightly better sounding vocal. Was it worth the extra effort? That’s going to be up to each person/situation.
The overheads on this track were also very bright and bordering on the harsh side. That said, the cymbals add a lot of energy to the mix so one thing I struggled with was keeping them bright and high in the mix, without making them entirely overbearing and harsh. I tried e²deesser on the overheads with the Overheads mode, and found the release too slow for this track. However, then I tried the Stereo Mastering preset across the 2-buss, and with a very low Amount, the cymbals just gelled nicely. This was a situation where I couldn’t get the same results with Pro-DS.
Next up, I loaded up a slower acoustic mix I use frequently for test projects. This singer has a lower voice, but one that has a bit of a rasp that can be problematic. When it came to this particular vocal, I just couldn’t get the e²deesser to work as well Pro-DS. e²deesser would audibly breathe and pump on esses. I experimented with a few different Modes (the manual indicates that Voiceover has a slightly faster attack than Solo Vocal), but none were anywhere near as effective as Pro-DS.
The Eiosis e²deesser is one of the most impressively featured de-esser plugins on the market. It’s extremely flexible while maintaining ease of use, and I’ve only just begun to discover what it’s capable of. Having the ability to treat the siblant and voiced signals differently is just absolutely brilliant, as are the EQ options baked right into this. While there was some tricky material where I felt I got somewhat better results elsewhere, e²deesser could easily become my desert island de-esser do to the combination of great sound quality and the insane number of features.
Sound Quality - 4/5: Works extremely well on some material, but there can be audible pumping in some instances.
Ease of Use - 5/5: Everything is clearly laid out and intuitive.
Features - 5/5: Honestly, this probably already has too many features (in a good way).
Bang for Buck - 5/5: This is about the going rate for a top of the line plugin de-esser, and that’s exactly what you’re getting here.