By Steve Cunningham
Voice processors that combine a mic preamp, a compressor, some EQ, and a limiter have become quite popular. They give you all the tools you need to produce great voice tracks in one box, without all the connection hassles of using a mixer and outboard processing. In addition, the quality of the individual sections in a vocal processor is usually better than what you’ll find in a typical mixer and outboard gear combo, and the whole business is much easier and more convenient to use.
Having said that, describing the Manley Voxbox as just a vocal processor is a bit like saying the Porsche Boxster is a good transportation car. The all-tube Voxbox is not just good, it’s astonishingly good. In fact, every audio block in this product is astonishingly good.
The Voxbox incorporates several existing Manley designs into one package, including their mic preamplifier, optical compressor, and Pultec-style passive EQ. To this, they have added an optical de-esser with a limiter setting, and they’ve packaged it all with enough input and output connectors to make it flexible.
The VoxBox is designed and built by Manley Labs in Chino, California. For those of you not familiar with the Manley brand, the company has been building all-tube hi-fi and pro audio gear since the 1980s, and Manley equipment is used by some of the best recording studios and mastering houses in the world.
Manley products are not inexpensive, and with a list price of $4000, the Voxbox is no exception. But if your livelihood depends on the sound of your voice, then you need to read on—this device will improve your life.
SIGNAL FLOW
The Voxbox is organized in two sections. The first section contains the compressor and mic preamp, and I mention them in this order for a reason. The input signal is first sent through a passive rumble filter, and then through the INPUT attenuator. From there, the signal hits the passive 3:1 compressor, which relies on an opto-isolator to do its work. The compressor has no amplification stage, and hence adds no noise. The signal then goes to the mic preamplifier, which adds between 40 and 50 dB of tube-driven gain.
Normally you connect your microphone first to a preamp, and from there to a compressor. By reversing the order of these audio blocks, the Voxbox actually reduces the possibility of clipping the mic pre by feeding it a signal that is already compressed.
The second section of the Voxbox contains the EQ and the de-esser/limiter. The EQ is a passive 3-band design, with bands for low frequency boost, midrange cut, and high frequency boost. These are passive equalizers, with nothing but a capacitor, an inductor, a rotary switch, and a conductive plastic pot in the signal path for each of the bands. There are no tubes in the EQ, and definitely no ICs or transistors. Once again, the result is an EQ that adds no noise.
After the EQ, the signal hits the de-esser which is also passive, and uses another opto-isolator to control a passive notch filter. You can select one of four fixed “ess” frequencies set at 3K, 6K, 9K, and 12K, or you can select “LIMIT” which switches the notch filter out, leaving a full range 10:1 limiter engaged. The final stage is a tube line amplifier, which compensates for any loss created by the passive EQ circuitry and allows the Voxbox to drive long cables from its outputs.
Input signals can be at mic, line, or instrument level, with XLRs for the mic and line inputs, and a 1/4" jack on the front panel for instrument-level signals.
THE FRONT PANEL
The three rack space front panel is a full quarter inch thickness of brushed and blue-gray anodized aluminum, with a stylish VU meter embedded on the left side. Graphics and labels are engraved into the aluminum, and the entire unit looks very classy and professional. Trust me, you’ll like this one in your rack; it’s gorgeous.
All controls on the Voxbox are either positive acting toggle switches, smooth pots, or detented knobs that click into place with a satisfying “thunk.” The anodized aluminum knobs extend out about an inch from the front panel, so it’s easy to grab hold of them.
The front panel of the Voxbox is well thought out and features some very good ideas. The PHANTOM power switch is a locking toggle that must be pulled out before it can be switched—a particularly good idea for knuckleheads like me who occasionally forget to turn on the phantom power until after I’ve raised the fader. This particular bonehead move has cost me a set of tweeters in the past, and the locking toggle is a nice little reminder to help me avoid doing something stupid.
Another thoughtful feature is the rotary switch under the cool-looking VU meter. With it you can check your levels from the line input, the preamp output, or the EQ output, and you can also check the amount of compression or de-essing that you’ve dialed in.
THE REAR PANEL
The rear panel features connectors that are grouped into the same two sections mentioned earlier. All inputs and outputs are available on XLR and 1/4" jacks, except the mic input. Rightmost on the back is the mic preamp/compressor section with an XLR for connecting a microphone at mic level. Next is the line input, which will handle both balanced and unbalanced audio at a +4dBm level. The preamp out completes the first section. Interestingly, Manley claims that the 1/4" preamp output is the cleanest output for direct to tape recording, since it bypasses the output transformers.
The second section contains an insert input, which injects signal into the EQ section and is intended as a return for an external noise gate or other processor. Finally, there is the EQ output, which acts as the final output of the Voxbox.
Two Stereo Link connectors complete the rear panel connectors, allowing you to link two Voxboxes together and have their compressors and de-essers work in lock step. There are separate ground terminals for both circuit and chassis ground, enabling the Voxbox to function hum-free in a variety of studio wiring configurations.
UNDER THE HOOD
Of course I took the top off; what did you expect? I’m pleased to report that the inside of the Voxbox is as impressive as is the outside. This is obviously a hand wired and hand soldered product, and several signatures are visible on components that have obviously been selected and checked for performance. (I don’t know who Carolina is, but I’d like to thank her for checking out those capacitors.)
The PC boards are robust, with very thick traces that appear to be able to handle high current. The several input and output transformers are labeled as being custom-built by Manley, and it is obvious that the unit’s performance is in part due to the extensive use of shielding between the various sections. The construction of the Voxbox is all first-class.
THE VOXBOX COMETH
I usually work alone in my studio, and record voiceovers using a Lawson L47MP large-diaphragm tube condenser microphone. The Lawson goes into a Yamaha 02R console, and from there to both a Macintosh computer with a Korg 1212 I/O card, and to a Sony DAT recorder as a safety. The Korg card is connected to the Yamaha console digitally via an ADAT light pipe, and the DAT recorder gets its signal from the analog output of the console.
I like this setup, and it produces what I believed were good-sounding voice tracks. Then one day, not long ago, the Voxbox arrived and everything changed.
The first thing I noticed on opening the box was the unassuming spiral-bound owner’s manual. Unlike some, I enjoy looking through a manual before I fire up a new piece of gear. It helps me avoid potential pitfalls, and it also gives me a good indication of the manufacturer’s attention to quality and detail, or lack thereof.
The manual for the Voxbox was such a good read that I didn’t actually fire up the piece for an hour after removing it from the box! Besides describing the operation of the Voxbox, the manual is a good refresher course on sound engineering practices that are often forgotten today. For example, in the EQ section the manual reminds you to try cutting first rather than boosting. Good advice.
THE PREAMP IN USE
I began by connecting the Lawson to the Voxbox, and connecting the preamp out of the Voxbox to a line input on the console. I reset the console channel to flat, with no EQ, compression, or effects. I set the Voxbox compression to BYPASS and read a couple of low-key, “warm and fuzzy” spots.
The first thing I noticed on playback was not additional warmth, as one might expect using an all-tube channel. Rather I noticed a quality of additional “air” in the vocal sound, and a marked increase in presence. The track sounded as if the talent was standing very close to me, speaking softly into my ear.
Being aware of the “placebo effect,” I decided to A/B it with my existing setup by recording the left channel with the Voxbox and the right channel without. On playback my first impression held—the channel without the Voxbox sounded choked and brittle in comparison to the channel with the Voxbox. The improvement was impressive. Hey, this thing is good! Time to try the compressor and see what happens.
I flipped the switch on the Voxbox from BYPASS to COMPRESS 3:1, and I set the VU meter to show me the gain reduction. As I do, I’m thinking that the fixed 3:1 ratio may be too low, as I usually compress at about 4:1 with a high threshold, so I really only hit the peaks but I hit them pretty hard.
With the THRESHOLD control set at about 11 o’clock so it kicked in only on the plosives, I then recorded a hard-sell spot that really pinned the input meter. The level of the resulting track was somewhat tamed, but the compression was inaudible. So far so good. Cranking up the THRESHOLD to 4 o’clock, I recorded the same spot. The result was a thicker sound that held the VU meters steady, but there was still no pumping, and no increase in noise. Just a very in-your-face sound, yet uncolored.
Changing the ATTACK and RELEASE controls yielded the results I expected. While I preferred both to be set to MED FAST for VO tracks, I found that setting them to SLOW gave a good impression of an engineer riding a fader. This was useful to smooth out rowdy music beds, although you’d want a second Voxbox to process stereo (hint, hint, I want another one of these, please).
I was able to borrow both an ART PRO MPA tube mic pre and a Bellari tube mic pre to compare them with the Voxbox. Both the ART and the Bellari are competent voice processors that cost much less than the Voxbox. They both imparted a tube warmth to voiceovers, but they colored the sound of the tracks. On the other hand, the Voxbox sounded far more transparent. It did not really give the same kind of warmth, but it also didn’t color the tone of the mic. It seemed to have a more natural, open sound.
This is one of the primary differences between the Voxbox and other tube voice processors. Manley seems to strive for clarity, and a complete lack of noise and coloration. While the price differences are dramatic, the differences in the sound quality between the Voxbox and the other tube voice processors was even more dramatic. I suppose that once again proves that you really do get what you pay for.
Altogether I tested the Voxbox with a few different microphones, including the aforementioned Lawson. The Lawson microphone is wonderful—you can record two tin cans and a string with the Lawson and get a good track. But how would the Voxbox handle other mics that are a bit less wonderful?
EQ AND DE-ESS
My beat-up old AKG 414 condenser mic was my first “pro” mic, but it has a nasty hump around 3.5kHz that makes me crazy. In the past, I’ve tried to EQ the hump away with limited success. Voice talent ends up sounding either harsh or, with midrange EQ, hollow. It’s particularly brutal with women’s voices.
So I connected the AKG 414 to the Voxbox to give the EQ and DE-ESSER sections a workout. Just to make life interesting, I invited a female voice artist to come over and re-record a track we’d done earlier on my pre-Voxbox rig.
Using the MID DIP, I cut about 4dB at 3K, and I added just a bit of 16K using the HIGH PEAK. I set the DE-ESSER frequency to 6K and experimented with the THRESHOLD until the VU meter showed just a couple of dB of de-essing. Finally, I added a little less than one dB of low end at 300Hz to give her voice some meat.
The resulting track was better than the original done on the Lawson/02R setup. Her voice had more “air” and a nice silky sheen on the top end, but was not at all harsh or gritty. When she commented on how much smoother she sounded and asked what I’d done differently, I just pointed to the Voxbox. Her response was “That looks amazing... what is it?” It’s my secret weapon, that’s what it is.
The sound of the Voxbox’s EQ section is remarkable. Although I wondered if I could live with a cut-only midrange control, I ultimately found it more useful than the traditional boost and cut mid EQ. As a bonus, the cut-only midrange makes you think about how and when you use it, which usually results in a better sound.
OTHER NOISES
During the evaluation period I tested the Voxbox with more than just vocals. I recorded a guitarist with a Strat and a full pedal board, using the INSTRUMENT input on the front of the Voxbox. The guitar player began by telling me how much he hated his sound when plugged directly into a console. He really wanted me to mic up his amp. I politely refused.
Manley recommends against using the Voxbox’s compressor with electric guitars that are plugged in directly, since compression lowers the impedance of the INSTRUMENT input and can cause a loss of highs. They suggest using the DE-ESSER in its LIMIT mode as a substitute for the compressor, or using an effects pedal between the guitar and the INSTRUMENT input. We opted for the latter approach.
We ended up with a smoking rhythm guitar track with a lot of punch, and the guitarist told me it was the first time he liked his sound recorded directly. We achieved similar good results with his lead tracks, and also with a bass track played on a G&L 5-string bass. Speaking of bass, although I feared for my studio monitors when the bassist hit the low 5th string, the Voxbox took it in stride and reproduced those low notes cleanly.
FEELING FRISKY
By now you’ve gathered that I rather like the Voxbox. In fact, I like it so much that I’ve behaved recklessly with it. One of my current projects is narration for a video, and I send the client audio tracks for approval as we progress.
On a whim, I recorded the last voiceover using the Voxbox and a Shure SM-57, burned the CD, and shipped it off. I later asked the client his opinion of the voice work—you guessed it. He loved it, said it sounded good, and told me to keep up the good work. I can do that. I will not claim that the 57 sounded as good as the Lawson, but I will say that the old 57 never sounded as good as it does through the Voxbox.
SUMMARY
The Manley Voxbox is a well thought-out, well-built piece of audio equipment. Although it carries a hefty price tag, it’s a good value, especially if your livelihood depends on delivering excellent voice tracks to your clients. I looked hard to find something I didn’t like on this unit, but I’ve come up empty.
Like many of us, I have a fetish for really good audio gear. While I can’t always afford the best, there are times that I make whatever sacrifices are necessary to own an exceptional piece. With the Voxbox, this is clearly one of those times. My clients have commented on the improvement in my sound, as have other voice actors. I really have no choice here. I’m not returning this unit—I’m writing a check.
♦