Esspresso review – Klevgränd Produktion add a de-esser to their iOS effects app line – and win a prize for ‘best app name ever’

Download from iTunes App StoreEsspresso logo 1I’ve reviewed Vandelay, SquashitSvep, RoverbEnklKorvpressor and, most recently, PressIt from Klevgränd Produktion previously on the blog and, in each case, was hugely impressed with the creative or corrective possibilities that these apps provide. They are also very very competitively priced and any of these would be worth including within your iOS audio app collection. Oh, and if you work on a desktop system also, many of these iOS apps are also available as AU/VST plugins…..

Today has seen a new release in the Klevgränd iOS app/desktop plugin product line; Esspresso and, as the app is a de-esser (essentially a corrective processor that attempts to reduce any sibilance caused by ‘ssss’ or ‘tttt’ or ‘sshhh’ or ‘zzzz’ sounds in your vocal recordings via some frequency selective compression), this has to be one of the cleverest app names I’ve seen in a while :-)

Esspresso; snake skin shoes never sounded so sibilant-free :-)

Esspresso; snake skin shoes never sounded so sibilant-free :-)

If you have used any of the Klevgränd Produktion apps listed above, then you will have some idea of the format to expect with Esspresso. All there apps bring a combination of a rather stylish, often novel, and streamlined user interface, a focused feature set, Audiobus and IAA support and easy (and generally very intuitive) operation. Esspresso doesn’t disappoint in any of those regards.

And, as with the other apps in the product line, it is modestly priced (under iOS, the launch price is UK£5.99 (US$7.99) while the VST/AU plugin is listed at US$49.99). The iOS app is a 3MB download and requires iOS7.0 or later. This is an iPad only app and, in my own testing this morning, worked very smoothly via Audiobus or IAA under iOS9.0.2 on my iPad Air 1 test system.


If you are relatively new to the wonderful world of multi-track recording (and no shame in that), you might still be getting your head around audio processing tools such as compressors or noise gates or EQ or effects such as reverb and delay. A de-esser is perhaps a more specialised (niche?) tool but most modern desktop DAWs will include a de-esser plugin in some form or other.

You can think of a de-esser as a ‘sibilance controller’. Sibliance is the ‘ssss’ noise that some singers make when pronouncing ‘s’ sounds and, depending upon their microphone technique and the microphone used, it can add some very distracting ‘sssss’-style noise and distortion to what might otherwise be a perfectly good performance. Some singers are also more prone to this than others (for example, Freddie Mercury tended to generate quite high sibilance levels apparently).

The app is supplied with a few presets to get you started but, once you have a grasp of the clever control set, it is very easy to configure and fine-tune for your specific vocal track.

The app is supplied with a few presets to get you started but, once you have a grasp of the clever control set, it is very easy to configure and fine-tune for your specific vocal track.

Of course, if your voice and pronunciation are a bit ‘spitty’ on the ‘s’ sounds, there’s not a lot you can do physically. In the general mayhem of a one-off live context it is perhaps less of an issue but, on a record, where a listener may play the same recording many times, it is, at best, distracting for the listener and, at worst, can simply make a vocal performance unusable within a mix.

This sibilance tends to be focused in the higher frequencies so you can try to reduce or remove it with a little bit of tightly focused EQ. However, without resorting to some automation so the EQ is only applied when the sibilance issue occurs, you can end up robbing your vocal sound of some of its natural high-end presence.

De-essers were designed to provide a somewhat more sophisticated solution to this issue and, actually, there are a number of different approaches that these corrective processors can adopt. However, in essence, what the processor does is temporarily reduce the high frequency element of the sound but only when the sibilance itself is present. One way of thinking about this is to see it as a frequency-dependent form of compression; the compressor only kicks in to reduce the level when there are high signal levels in the sibilance frequency range and the compression only occurs within that frequency range. In most decent de-essers, therefore, the user gets (a) some control of the frequency range to target and (a) the amount of compression to then apply.

Make mine an Esspresso

Esspresso does both of these things (although it also does a little more as we will see below). In yet another triumph of minimalist-looking design, in the single screen of controls the app offers, you get a large waveform display with low frequencies to the left and high frequencies to the right. The display is, however, split into an upper portion that shows the frequencies of the incoming audio (pre-processing) and the lower portion showing the outgoing frequency response (post-processing).

If you visit the Klevgränd website, you can download the concise - but helpful - manual for Esspresso.

If you visit the Klevgränd website, you can download the concise – but helpful – manual for Esspresso.

In Esspresso-speak, the upper portion of the interface is termed the Detector (it’s where the app is ‘listening’ to the incoming audio and the user defines the conditions under which the app detects where sibilance might be an issue) and the lower portion is termed the Suppressor (where the user defines the frequency range any corrective compression is applied over and the level of that compression).

'All' mode applied compression to the entire frequency range.....

‘All’ mode applied compression to the entire frequency range…..

In terms of controls, in the upper portion, on the right you can adjust the Sensitivity (essentially, this adjusts the input gain) while on the left (although you can stretch this over the whole of the frequency range in the upper portion of the display if you wish but this would make your de-essing rather non-selective) you define the frequency range over which the app should ‘look’ for high input levels. This is shown via the ‘highlighted’ window area and you can move and re-size this to control the exact range of frequencies Esspresso is looking at. Usefully, if you toggle the ‘S’ (solo) button on (located on the left beneath the Sensitivity slider), then you can monitor just this frequency range of the input signal. This makes it much easier to identify what the problematic area might be for your particular input source.

Within this frequency window there is also a horizontal line and this can be dragged up/down to set the threshold at which any de-essing is to be applied. If you just want the de-esser to be applied to the most extreme ‘sssss’ sounds, then you set a high threshold (so the peak levels just exceed the threshold line). However, if you want a stronger de-essing effect (to tame things more assertively), then a lower threshold setting can be used.

In 'low' mode, the compression is only applied to those frequencies above the user-defined range;low frequencies are left untreated for a more natural sound.

In ‘low’ mode, the compression is only applied to those frequencies above the user-defined range;low frequencies are left untreated for a more natural sound.

In the lower portion of the display – the Suppressor section – you can configure a number of things about how compression is applied once the Detector section has detected something above your threshold and within the frequency range set. So far then, this is so standard de-esser practice but the Supressor actually gives you more control than many de-essers.

First, you can choose between three ‘frequency band options’ via the selection buttons located bottom left; all, band or low. In ‘all’ mode, once the Detector say some de-essing is required, in fact the Suppressor applied its compression across the full frequency range of the audio (in effect, the app is acting as a more conventional compressor even if it is only being triggered by a specific frequency band set in the Detector).

In ‘low’ mode, you can set a frequency and, below that, no compression is applied (a kind of ‘low pass’ compressor), even when the Detector says the sibilance threshold has been exceeded. However, compression is applied above that frequency.

However, perhaps the most useful mode is ‘band’. In this mode, as in the Detector panel, you can define the frequency range over which compression is applied in the Supressor. You could, of course, match this frequency range in the Suppressor with that used in the Detector… but you don’t have to. That’s actually a very good thing and means you can choose to be very selective in the frequency range used in the Detector (and focus in on the most obvious ‘ssss’ frequencies for the detection stage) but then actually apply the compression over a somewhat wider frequency range within the Suppressor.

Band mode gives you the most control over the de-essing.

Band mode gives you the most control over the de-essing.

Finally, the Supressor’s highlighted frequency range also has a horizontal line that you can swipe up/down. This sets the amount of compression to be applied….   drag it down for higher compression ratios and a ‘stronger’ de-essing effect.

So, Esspresso is very aptly named….   it deals with ‘ssss’ sound by (frequency dependent) compression…. top notch app naming by a development team that obviously has a sense of humour…. and check out the demo video below for confirmation of that; very clever :-)

In use

I’ve only had a few hours to explore Esspresso so far but, on the whole, it seems to be very solid in operation. I did the majority of my testing via IAA and, as with a number of IAA apps, when testing within my usual IAA host (Cubasis), I did occasionally have to restart Esspresso after it and Cubasis seems to lose touch with each other. This is something I have experienced with other IAA apps so, whether it’s an Esspresso issue, and IAA issue or a Cubasis issue, I couldn’t positively say. It is, however, a minor issue…..

In audio/processing terms, the app worked well when used within Audiobus but, on my test system at least, I did seem to have some graphical issues (no waveform display?). If this is a general issue then I’m sure Klevgränd Produktion will, in their usual fashion, address it in a prompt fashion. Even as it stands, it certainly wouldn’t stop me using the app.

Esspresso worked happily within Audiobus under iOS9, although I did experience some graphical issues.

Esspresso worked happily within Audiobus under iOS9, although I did experience some graphical issues.

In terms of the results the app can produce, however, I’ve absolutely no complaints. The interface is beautifully designed and the way you configure the controls just make so much sense. Once you have grasped the basics, the ability to fine-tune your de-essing using the app’s key controls is really very good indeed. Having tried the iOS version, I’m now also very tempted by the VST/AU version :-)

This is, of course, a bit of a niche app and it might not be the first processing app you would buy if you were just starting your iOS music app collection. However, if you record any amount of vocals – and work with different vocalists on a regular basis – this would most certainly be a very useful corrective processing tool to have at hand.

The app worked very well via IAA within Cubasis.

The app worked very well via IAA within Cubasis.

Some vocalists are just very sibilant and, while there are things you can do while tracking to reduce the problem at source, inhibiting a singer by telling them to stop hissing is a sure-fire way to make them self-conscious and kill their performance. Do what you can to (unobtrusively) reduce the amount of ‘ssss’ that gets recorded but, if you have to tame it a little more after the fact, then Esspresso is a powerful tool for doing just that.

In summary

Like SquashIt – Klevgränd Produktion’s multi-band compressor app – Esspresso is perhaps quite a specialist processor and will appeal most to those recording vocals. However, also like Klevgränd’s other apps, while it does a very specific job, it really does do it rather well. There are not many stand-alone de-esser apps out there on the App Store (although there are a few built into other apps such as VocaLive) but, even if there was some obvious other competition, Esspresso would still be hard to beat.

If you record vocalists, forget today’s espresso and buy an Esspresso instead…. Niche app, but for what it does, highly recommended. Oh, and here’s hoping that Klevgränd also have producing iOS AU versions of their various apps near the top of their ‘to do’ list… :-)


Esspresso from Klevgränd produktion on Vimeo.

Be Sociable; share this post....


    1. Thomas Renhult says:

      Does Esspresso work on iPad 2 or do I need a newer model?


    Speak Your Mind