Aphex's Aural Exciter was the original psychoacoustic enhancer.
Psychoacoustic enhancers are a frequent topic of enquiry here at the SOS offices. Hugh Robjohns provides the answers to some of the most common questions.
In the early years of tape‑based recording — let's say the '50s and '60s — equalisation and compression, a little delay and some reverb were all the signal processing that was needed. So many of those early recordings have fantastic clarity, detail and sheer acoustic involvement! Simple but careful mic placement, short signal paths and, frequently, straight‑to‑stereo recording techniques were probably the key to maintaining the detail and clarity — keeping it simple also kept it sweet.
There is nothing to stop modern recordings being made the same way, and those that are invariably sound fantastic — to my ears at least — but the current fashion is for more elaborate and, dare I say, more artificial sounds and performances. Technology has advanced dramatically over the last few decades, giving us ever‑more‑complex console signal paths, multitracking and overdubbing techniques and the ability to compile tracks from numerous takes. Although these technologies have made possible creative techniques which were barely dreamed of 30 years ago they have, in the process, made it all too easy to lose some of the essential sonic quality of the recording itself.
Complaints of lacklustre, dull‑sounding mixes that fail to excite are commonplace, and consequently, many solutions to this particular problem have appeared. The usual solution is to reach for the 'psychoacoustic processor' to bring back the 'edge' and 'clarity' to the mix (or elements within it) and in this FAQ article I have tried to answer many of the most common questions concerning these devices we receive at Sound On Sound.
Q. When was the psychoacoustic processor invented?
BBE's enhancer designs use phase realignment and dynamic equalisation.
Aphex were the first company to introduce the concept of psychoacoustic processing, with their Aural Exciter back in 1975. Originally, these devices were only available for hire — charged out at $30 per minute of finished recording time — but as their popularity grew, Aphex decided to sell both the professional models and the subsequent (and far more affordable) Type B and C units. The term 'Aural Exciter' is registered to Aphex and so functionally similar units from competing manufacturers are generically known either as 'psychoacoustic processors' (which can be something of a mouthful!), or more commonly just as 'enhancers'. Broadly comparable products are now available from the likes of Aphex, BBE, Joemeek, SPL, and Behringer to name but a few; most are analogue signal processors, although a few digital units are now beginning to appear.
Q. So what is a psychoacoustic enhancer?
The Joemeek VC4 employs a unique enhancement process based on frequency‑dependent compression.
Perhaps the best way to answer the question is to look at what the device is aiming to do. As I mentioned in my introduction, with modern recording techniques and equipment, recordings can tend to become a little dull and lifeless or, put another way, to lack presence and sparkle, particularly if the recording process involves multiple stages of recording and post‑production. This problem is often blamed on absent or diminished high‑frequency harmonics — the essential components of original sound which embody the 'life' and 'air' of the instrument or voice. Enhancers, in all their various forms, are designed to repair, restore or replace this part of the audio spectrum, recovering the clarity and detail which was previously perceived to have been lost.
A simplistic, though easy, solution to this problem is just to apply high‑frequency equalisation, with the intention of boosting the amplitude of whatever upper harmonics may remain present in the signal. However, this is not usually a very practical answer on its own, because this kind of amplitude processing increases and emphasises any high‑frequency noise in the original signal. It is also highly likely that it will distort or otherwise upset the overall tonal balance of the recording in a detrimental way.
Over the years three main enhancing techniques have been developed. The first and original system is Aphex's, which relies upon the regeneration of the supposed harmonic content of the original signal. Other manufacturers use a number of variations on this principle, the precise means of generating or restoring programme‑related harmonics being designed to avoid infringing Aphex' original patents.
For example, BBE's Sonic Maximiser system uses a combination of dynamic equalisation and multi‑band phase realignment. The SPL Vitalizer also uses dynamic equalisation, but combines the modified signal (using a technique that is a closely guarded secret) with the original in such a way that avoids compromising the mid‑band detail and clarity, at the same time as introducing some phase‑shifting technology with commensurate benefits to the high‑frequency region.
Q. How does the Aphex 'Aural Exciter' work?
SPL's Vitalizer enhancers offer two types of bass enhancement as well as dynamic high‑frequency equalisation.
The Aphex solution is complex in detail but, essentially, the user tunes in and isolates a specific mid‑ and/or high‑frequency portion of the original signal which the processor then deliberately sets about distorting! This non‑linear process generates a series of musically related artefacts which consist mainly of second harmonics (with a small amount of third and higher harmonics). These 'regenerated' harmonics can then be added back to the original sound in the desired amount to provide the required 'sparkle' and 'air'. A further level of sophistication is the introduction of a degree of compression to the harmonic signal to strengthen and maintain a consistent effect regardless of the level of the original material, and some frequency‑dependent phase‑shifting.
While most machines marketed as enhancers are intended to achieve similar end results, the means they use to get there can be very different indeed. When used carefully, the audible effects of most enhancers are remarkably similar and the real differences often only become noticeable when extreme settings are employed.
The fundamental idea is, simply put, to restore or recreate the initial harmonic spectrum which was (presumably) present at the outset in the source, if not the recorded signal. As these newly generated harmonics are derived directly from the spectral content of the original signal, the system adds minimal noise and does not (assuming sympathetic application) alter the tonal balance unduly — significant improvements in both respects over the simple equalisation approach. The operational controls of an Aural Exciter are typically limited to Frequency and Mix, which set, respectively, the lowest fundamental frequency for harmonic generation (usually somewhere between 3 and 8kHz) and the amount of harmonic signal added in to the composite mix.
Q. Are all enhancers the same?
Not quite. While most machines marketed as enhancers are intended to achieve similar end results, the means they use to get there can be very different indeed. When used carefully, the audible effects of most enhancers are remarkably similar, and the real differences often only become noticeable when extreme settings are employed.
Only the Aphex system (and its immediate imitators) are true 'Aural Exciters' which go to the trouble of generating the missing or diminished upper harmonics. Both the BBE and SPL systems (and their ilk) rely more on applying carefully designed dynamic equalisation processes which typically combine a modest recess in the mid‑band with a slightly boosted treble region. In 1933, Fletcher and Munson demonstrated that the sensitivity of the human hearing system to high‑pitched sounds drops with decreasing sound pressure (volume). The idea of dynamic equalisation, therefore, is to bolster the upper harmonics and high‑frequency detail in the signal only when its level falls. An advantage of this kind of dynamic equalisation is that it only affects lower‑level audio signals and so does not compromise headroom in any way, yet enhances the perceived clarity, presence and detail in the recording at all levels. In essence, dynamic equalisation systems are designed to emphasise, rather than synthesize, upper harmonics.
Q. What about low‑frequency enhancement?
Many systems offer some provision for enhancing the low‑frequency spectrum, most commonly by further dynamic equalisation which raises the relative amplitude of any low‑frequency content as the overall volume drops. In concert with the HF boost and mid‑range depression, this LF boost results in the classic 'smiley' equalisation curve — a dynamic 'Loudness Filter' if you like — which is widely recognised as producing a pleasing and appealing sound characteristic. Just ask any loudspeaker manufacturer!
The SPL Vitalizer system provides a mechanism to adjust the bass sound between a deep, wide, expansive bass end, or a tight, well‑defined, punchy LF. This is achieved, in part at least, by altering the relative phase at which the dynamically equalised low‑frequency region is added back into the main body of the sound.
Enhancers can be extremely useful tools but, as with anything else, over‑use quickly negates any benefits, and the ear gets used to the effect extremely quickly.
Some of the Aphex Aural Exciter systems incorporate a bass‑enhancement function under the marketing name 'Big Bottom', which is claimed to synthesize subharmonics one and two octaves below the lowest fundamental frequency in the audio programme material. The generation of subharmonics presents a similar range of advantages over low‑frequency equalisation as its high‑frequency counterpart. Identifying the lowest fundamental and generating sub‑octaves is a relatively noise‑free system, and does not risk eating into the available headroom in the same way as indiscriminate low‑frequency boosting equalisation, which will amplify rumbles, traffic noise and any other subsonic rubbish along with the wanted low‑frequency musical information.
Q. So are enhancers just dynamic equalisers by another name?
No: dynamic equalisation, by itself, is not really enough to restore the brilliance and clarity which is required, although it does go a long way towards it. Generally, other signal‑processing elements — usually frequency‑dependent phase shifts — are employed to strengthen the effect, in an attempt to realign the relative timing of low‑, mid‑ and high‑frequency components in a complex sound. This has a profound affect on transient signals, in particular, which play a very important role in our recognition of sounds and in the determination of their spatial positioning.
The various BBE Sonic Maximiser systems employ exactly this kind of relative phase alignment, which they claim is essential to compensate for the inherent phase shifts associated with conventional equalisation and filtering circuits. This particular system divides the audio spectrum into three bands before applying corresponding frequency‑dependent delays, low frequencies being delayed the most and high frequencies the least. This arrangement creates opposing phase delays to those encountered during passage through normal analogue signal processing chains as well as in conventional loudspeaker reproduction. The aim of these compensating frequency‑dependent delays is to mirror them and thus cancel out this particular source of sonic distortion. BBE also claim that their 'phase realignment' process helps to restore the true shape and accuracy of any original transient waveforms and their overall amplitude envelope shapes
Another variation on the theme of dynamic equalisation has been adopted in the Joemeek VC4 enhancer. The design of this particular device applies a form of dynamic equalisation in a unique way. Effectively, an appropriate portion or region of the mid‑ and/or high‑frequency spectrum of the input signal is isolated and subsequently compressed, before being added back into the original. A bandwidth control determines the breadth of signal to be processed, a Drive control sets the amount of compression applied, and a control labelled Enhance adjusts the amount of compressed signal added back in. The end result is a form of dynamic equalisation: low‑level signals are more strongly affected by the addition of the compressed component than full‑amplitude signals, and so the detail and clarity of these signals is enhanced.
Q. What is the best way to connect an enhancer to my mixer?
Enhancers are used much like any other external effects processor, and can be connected either in an effects loop of the mixer or in an insert point: the best approach depends on the specific application and the facilities available in the particular unit. Some engineers claim it is best to drive the processor from an auxiliary send (or two for stereo) and return the processed signal to a dedicated effects return or spare channel. However, if used in this way it is imperative to ensure that only the enhanced signal is returned — you don't want the original to be returned as well in this application, since the original signal will already be present in the mix.
Many enhancers are designed to be used purely as insert processors, either in individual channel inserts, or group or main mix buss inserts, since the recombination of processed and direct sound can only be adequately performed within the unit itself. It is worth paying attention to optimising the input and output signal levels in this situation, as many console insert points do not operate at standard +4dBu line levels.
Q. How should I use an enhancer?
Enhancers can be extremely useful tools but, as with anything else, over‑use quickly negates any benefits, and the ear gets used to the effect extremely quickly. Personally, I look on enhancers as a last resort — only to be used if all else has failed — although I admit that this is a highly personal viewpoint! It is always far better to take as much care as possible to capture a crisp, bright, clean recording in the first place. Elements of the signal lost at the start of the recording process, through sloppy mic technique or poor equipment, can never be adequately replaced. The concept of 'fixing it in the mix' is completely bogus — it never, ever, works and is always a bodge! An Irish colleague once described this attitude as being the equivalent of "gold plating a turd — you can make it as shiny as you like, but it's still a lump of shit!"
However, we are rarely in full control of events and the perfect recording is still to be made. It may be, for example, that a specific instrument sounds a little dull for any one of a dozen uncorrectable reasons; or it may be lacking in presence when compared to other instruments; or it might just not have enough of an edge to cut through the mix sufficiently. In these kinds of instances the use of an Enhancer to add a little extra definition, presence and clarity to a specific instrument (or small subgroup of related instruments) can be a real blessing and often makes all the difference. Used as only a small component process within the final mix in this way, the effect is usually extremely benign, as it does not generally draw undue attention to itself and the results can be immensely effective and natural.
However, I would strongly advise against recording an enhanced signal directly — this really is a process which should be applied carefully in post‑production where its effect can be assessed more accurately in the context of the final mix.
A further point to consider, this time in specific relation to the genuine Aphex Aural Exciter process, is that, since this is a system intended to generate additional second and third harmonics, its use with other devices that perform similar functions could be unwise. An example might be applying the Aphex process to a heavily distorted guitar track — the result of which can be pretty hard to listen to! Having said that, of course, there are those producers and engineers who actively seek to obtain precisely this kind of hard‑edged and aggressive sound. Enhancers which rely more on dynamic equalisation are far less likely to suffer from this kind of effect, although they should still be used with care in this kind of situation, as a distorted guitar possesses a great deal of high‑frequency harmonic energy.
Q. What about enhancing a complete mix?
Take care! Most problems occur when a stereo enhancer is strapped across a complete mix and, if it is not adjusted with extreme caution, processing a complete mix quickly leads to a very unnatural, tiring and fatiguing sound, albeit a superficially exciting one! As I mentioned earlier, the ear quickly becomes used to the accentuated sound of an Enhancer, in much the same way that inappropriate equalisation can quickly sound 'normal' in isolation, so it is wise to apply any global enhancement processes sparingly and cautiously.
The best advice, as with EQ, is to make frequent use of the Bypass button to compare the original and processed sounds. Identify the failing in the original sound which you are attempting to address and then check whether your 'enhancement' is really benefiting the mix, or is actually introducing an unwarranted level of artificiality to the sound. If in doubt, it is far better to back off and apply too little than suffer the fatigue of excessive processing.
The choice between the original Aphex Aural Exciter, or one of the various generic enhancers is, inevitably, a very personal one. If used sparingly, all sound very similar and are equally capable of adding a useful degree of extra clarity and detail. Unfortunately, the Aural Exciter has been overdone to such an extent on so many records that it has become an instantly recognisable effect in its own right. The various offerings based around dynamic equalisation and phase‑realignment provide a different kind of sound enhancement, but one which is every bit as effective when used judiciously.