Focusrite Green Voicebox

Many a budding engineer aspires to Focusrite equipment, but until recently, its cost has made it the exclusive province of the professional. The Green range aims to change all that, while still retaining the legendary Focusrite quality. Paul White gives it the green light.

The Focusrite Red range of signal processors will be familiar to just about anyone who has worked in top‑name studios, but few project studio owners are able to afford the premium prices that go with them. From their massively engineered, red‑anodised front panels to the design excellence of their hand‑built circuits, these units ooze quality from every pore and socket — but if you can't afford one, you can't afford one, no matter how good it is.

Recognising that there are lots of quality‑conscious project studio owners who would like to own Focusrite equipment if it could be offered at a more affordable price, Focusrite went back to the drawing board and came up with the Green range. Obviously, green anodising isn't significantly cheaper than red anodising, so more had to change than simply the colour — and as you can see from the photograph, it has. Focusrite wanted their new range to make a visual as well as an aural statement — which is why the front panels of the new units look as though they've been partly assimilated by the Borg! Non‑Star Trek Next Generation fans can pretend they look like an aerial view of a golf course with Trident nosecones poking out of the bunkers.

It appears that the front panels and cases of the new range are all cast in one piece, though there could be an invisible join somewhere under the black crackle paint, for all I know. The double 'f' Focusrite emblem is embossed on the covers, and the ends of the cases are curiously rounded. The green front panels appear to be mounted behind the black sculpted front covers, so that the control knobs and buttons peep through. All the knobs have a conical profile and a tactile, rubbery feel, and the buttons are made of a similar material.

To cut component costs, Focusrite have dispensed with transformer coupling and replaced it with a properly‑designed, electronically‑balanced alternative. To cut manufacturing costs, the PCBs make extensive use of surface‑mount components, enabling much of the manufacturing process to be automated, without sacrificing the benefits of Focusrite's original circuit topology.

Colour Box

Aptly named the Voicebox, the first unit we have for review from the Green range combines a mic preamp, a compressor, a de‑esser, an expander, and an equaliser, in a single‑channel package which occupies one unit of rack space. Like all the other voice channels and voice processors that have recently appeared on the market (two of which, the LA Audio Classic Channel and the TL Audio VP2051, are also reviewed in this issue), its aim in life is to enable you to record high‑quality vocal sounds onto tape, without having to go via a mixer. No valves or 'vintage' gimmicks are employed in the Voicebox — just really good circuit design. For example, although there are three stages of dynamic processing, the signal only passes through a single, class‑A VCA. Two further VCAs are used in the side‑chain circuit, but the audio path is short and sweet. A further special chip is used in the expander side‑chain, due to the fact that expander thresholds are invariably much lower than compressor thresholds.

As you'd expect, no external power supply is required for the Voicebox, and the XLR inputs and output are fully balanced and grounded in accordance with the current AES (Audio Engineering Society) recommendations. Two rear‑panel jacks provide the options of remote muting and stereo linking of two units.

Design Focus

At the start of the signal path is a very low‑noise mic amp, providing from 10dB to 60dB of gain, with switchable phantom power, a phase switch, a mute button, and a switchable 75kHz high‑pass filter. Each of the buttons has an integral status LED, and a 5‑section LED level meter peeps through its own front‑panel slot above the mic amp section. If an external mute control is applied to the unit, the red mute LED comes on.

Also sharing the mic amp is the NR (Noise Reduction) control, which sets the expander threshold. The expander, which operates with a 2.1 ratio, has a strangely‑calibrated knob, in that it shows the worst‑case noise level below 0dB on the meter; in other words, it shows how much residual noise you might expect to have left when the expander is operating. Though the Voicebox itself is about as quiet as theory allows it to be, the use of compression will reduce the signal‑to‑noise ratio of the input signal by the same number of dBs as the amount of gain reduction being applied. Careful use of an expander can minimise this, by applying 'soft gating' during pauses.

As compressors go, this one has very few controls — just three knobs and three buttons. According to the graphs in the manual, the compressor has a conventional hard‑knee characteristic, with user control over the Threshold, Ratio and Makeup gain being provided. The attack and release times are pre‑optimised for vocal use, but there's an auto‑release function which can be switched in to deal with material that has unusual characteristics. This works by progressively slowing the release time if the input exceeds the threshold either by a large amount or for a long duration. Because compression reduces the level of signals exceeding the threshold level, a make‑up gain control is necessary to restore the peak level. The remaining buttons bypass the compressor section and switch the level meter to read gain reduction.

Following the compressor is a de‑esser, and though the controls and side chain for this are quite separate from the compressor, the same VCA is used to restrain the signal gain. In effect, a de‑esser is a compressor that is particularly sensitive to high‑frequency sibilance, of the kind produced by some vocalists when singing 's' and 't' sounds. When these sounds are detected, the de‑esser pulls down the gain, to make the sibilant sounds less noticeable. A conventional Threshold control sets the level above which sibilant frequencies are treated, and a Bandwidth control determines which area of the audio spectrum is reduced in level. Simpler de‑essers compress the entire audio spectrum, which can lead to unnatural gain‑pumping effects. The de‑esser's Bandwidth filter is simple to use, but is actually quite complicated in terms of design. A 3rd‑order filter sets the low‑pass point, with a 1st‑order filter looking after the high‑pass point; as the bandwidth is increased, the lower limit slides down the audio spectrum as far as 1kHz. At minimum bandwidth, the filtering applies only above 10kHz. A 2nd‑order, fixed‑frequency filter operating at 1kHz is used in the side‑chain circuit to ensure that low frequencies do not trigger the de‑essing process.

On voices, the EQ is simply stunning in its ability to create the desired tonality while leaving the voice sounding perfectly natural.

Next in line is the equaliser, configured as three bands with variable frequencies. The low and high sections are shelving, while the mid‑range can be swept between 300Hz and 3.3kHz. Though this isn't a true parametric EQ, the bandwidth can be set to normal or Notch — the Notch setting has a very high Q, specifically for attenuating awkward spot frequencies. All three sections provide up to 18dB of cut or boost, though it is unlikely that much boost would be used in the Notch position, other than to identify problem areas or to create special effects.

Finally, there's an output level control, which has an associated overload LED to warn of impending clipping. A recessed yellow LED indicates that the unit is powered up.

Sounds Transparent

It would be difficult for the Voicebox to sound as distinctive as it looks, but it definitely rises to the challenge. However, while other vintage‑style processors try to be distinctive by what they do to the sound, the Voicebox is characterised more by what it doesn't do. You might think that the signal path on your mixer is clean and transparent, but next to the Voicebox, most mixers start to show their weaknesses. For starters, the Voicebox has loads of headroom and a very fast transient response, so everything comes out sounding smooth and natural. Abrasive sounds usually occur when circuitry can't respond fast enough to the input signal, or can't accommodate its peaks, but there's none of that here. The mic preamp behaves perfectly in every way and the expander proves to be so transparent that it's hard to tell it's on at all. The only thing you notice is the silence between sounds.

Checking out the compressor was also a revelation — vocal levels are controlled more positively than on just about any compressor I've tried, other than possibly a top‑end Aphex, yet there's little subjective evidence of processing, even at very high levels of gain reduction. What's more, the compressor doesn't seem to push the louder sounds away, as some compressors tend to do — even the most uneven performance is levelled surprisingly naturally.

The de‑esser, too, manages to clean up normal amounts of sibilance without causing the sound to dip or lurch — no doubt the clever filter arrangement is behind this. If you really over‑process, you can hear the sound becoming just slightly dull whenever the de‑esser is triggered, but when set up properly, it's wonderfully transparent.

Focusrite have a strong name in the EQ business, so I was expecting great things from the Green EQ section — and it didn't let me down. EQs normally come as 'gentle and musical' or 'firm and ruthless', but Focusrite have managed to create an EQ that really digs into a sound and changes it, yet without making it unnatural. Even when you crank on 10dB of top end, the sound just gets crisper. It's the same at the bass end — you can add as much warmth and punch as you like, but there's no lack of clarity or loss of mid‑range definition. On voices, the EQ is simply stunning in its ability to create the desired tonality while leaving the voice sounding perfectly natural. It might sound like magic, but again, it's down to well‑engineered circuit design and an appropriate choice of filter characteristics. When you boost top end using a cheap EQ and the sound becomes harsh, it's not the extra top that makes the sound rough — it's the circuit's inability to remain 'linear'.

Going Green

The Voicebox may be part of Focusrite's affordable range, but it is still quite expensive by comparison with other 'voice channel'‑type products. However, I think its performance justifies the price premium, and while there is an attraction in using tubes, FETs or photocells to recreate vintage effects, there are occasions when you want to feel that there is nothing at all between you and the original sound.

It's very difficult to set up the Voicebox to sound bad. (Indeed, to be cynical for a moment, part of the reason for the popularity of Focusrite gear may be that producers and engineers know that they can get great sounds without having to be all that precise when setting up.) It's as though the technology is on your side for a change! In fact, the only minor disappointment is that there's no line input on the Voicebox — it would be nice to be able to use the compressor or EQ when mixing, as well as when recording. In theory, you could use the mic input with the pad switched in, but a dedicated line input would have been more flexible.

It would be possible to run tests on this unit to verify its audio bandwidth, its headroom, its phase response and its transient response, but all too often manufacturers manage to come up with an impressive technical specification while still failing to deliver in the 'ears' department. Here, the impressive spec is just a starting point — the equipment is designed first and foremost to satisfy the ears, and it doesn't disappoint.