You are here

Antares AVP1

Vocal Producer By Hugh Robjohns
Published May 2002

Antares have packed their acclaimed Auto-Tune and mic modelling processes, along with dynamics and EQ, into this 1U rack unit.

Producer packages and voice channels of one form or another abound these days, but this neat little unit from Antares takes the idea in a new and rather interesting direction. The Antares Vocal Producer — more easily referred to as the AVP1 — is a clever integration of some of the company's existing products, plus a few other 'fairy dust' elements to make a surprisingly powerful and effective, but relatively easy-to-use voice processor. Before going any further it would be worthwhile my pointing out that, unusually for a voice processor, the AVP1 does not contain a mic preamp stage — it is a line-level device and must be used either as an insert to a mixer channel or effects loop, or with a separate stand-alone mic preamp.

Antares AVP1 Vocal Producer.Photo: Mike CameronAlthough it doesn't contain a mic preamp, there is a surprisingly large amount of fun technology! It starts with Antares' mic modelling and valve emulation, followed by Auto-Tune real-time pitch correction. Then there is a variable-knee compressor, an expander/gate, a tuneable de-esser, a two-band equaliser (which is far more flexible than it sounds), and an automatic double-tracking facility (using Auto-Tune to create the second voice). Most parameters can be accessed and adjusted with only a couple of button presses, and the whole shebang is fully programmable. It is also equipped with a broad range of factory presets designed to complement particular instruments and voices, as well as different musical styles. These serve as excellent starting points for inveterate twiddlers, and offer pretty good instant solutions for those who would rather get on with the music making. Most parameters can be controlled via MIDI continuous controllers too, for automation from a sequencer.

As with Antares' other rackmount boxes, this is a very compact 19-inch unit, measuring roughly 125mm from front to back. On the rear panel are three unbalanced quarter-inch jack sockets: one for the input and two for the outputs (main and double-tracked — of which more later). These are all at line level and seem to operate at somewhere around a nominal -8dBu (ie. roughly -10dBV). Having said that, this is a digital processor, so perhaps it would be more meaningful to tell you that peak level (0dBFS) equates to roughly +4dBu. Optimising the input level is a little tricky as, although Antares have provided a digital level trim, there is no analogue input level control — so you have to get the levels pretty much right from your mixer or other signal source.

The rear-panel analogue audio I/O is unbalanced, and it is joined by MIDI In and Out, a footswitch input and the inlet for the external power transformer.The rear-panel analogue audio I/O is unbalanced, and it is joined by MIDI In and Out, a footswitch input and the inlet for the external power transformer.

There are two MIDI sockets (In and Out) for remote control and user data transfers, and a quarter-inch footswitch socket, again for remote control of various functions. The AVP1 is clever enough to detect whether the footswitch is of the normally-closed or normally-open type when it powers up, and behaves appropriately thereafter.

Unfortunately, the Vocal Producer does not contain its own power supply (boo, hiss!), but an external wall-wart is supplied to deliver 9V AC to a coaxial socket at the rear of the unit. Even more unfortunately, there is no front-panel on/off switch, so you will need to be able to unplug the wall-wart (or reach its wall socket switch) when you want to turn the unit off. Yuk!

Hooking the AVP1 into a system is pretty straightforward, and the handbook offers several alternatives. Typically the unit would receive its input from the insert send of a mixer channel handling an appropriate source (single voice or monophonic instrument), the AVP1 main output being returned through the same insert point. If the double tracking facility is being used, then the secondary output would have to be returned via a second channel. Instead of insert sends, it is also possible to use a channel direct output or even an auxiliary send to drive the machine, with the returns being routed back into the mixer on separate channels or returns.

In Control

Moving around to the front of the AVP1, the control panel is well laid out, with six distinct control sections and a large two-row, backlit LCD. Each section has its own illuminated bypass button (labelled On) surrounded by four or five other buttons to access various parameters of the corresponding signal processing. Selected parameters are detailed on the LCD and adjusted with the control knob and page buttons. A global bypass mode is also available via the output gain section (see below).

The six signal-processing sections are (from left to right): Master (input and system configuration), Mic Modeler, Auto-Tune, Compressor/Gate, De-esser, Equaliser/Output. The last stage is actually subdivided into three, with two bands of EQ, a double-tracking facility, and an output level control with overall bypass.

The Master section is used to configure the machine, and whereas all the other sections have On buttons as their central control, this section has a button labelled Setup. This is surrounded by four other buttons: left and right cursors for navigating the LCD menus; a Page button to access any additional menu pages; and Preset and Save to recall and store presets. There are 35 factory presets which may be overwritten with user programs as desired. The original factory programs can be reinstated individually or globally if required.

An input level meter is provided in the form of a vertical lozenge-shaped display which contains five LEDs covering a 30dB range. A more detailed input level display can be accessed on the LCD and similar-resolution gain-reduction meters can also be accessed from the Compressor/Gate and De-esser sections. Strangely, there is no option to monitor the output level in this way, although there is a lozenge LED meter in the output section of the unit.

The mic modeller is based on the company's previous AMM1 stand-alone unit, but offers a diminished collection of source and modelled microphone types. There are a handful of specific source mics which include common stage and budget studio mics with denoted models from Shure, Audio Technica, Rode and CAD. There is also a variety of generic source mics such as handheld and studio dynamics, small- and large-diaphragm condensers, and wireless mics. The modelled options are all generic and include handheld and studio dynamics, two small- and three large-diaphragm condensers, a trio of drum mics and a telephone emulation. High-pass filtering options are provided for both source and modelled mics to allow adjustment of proximity effects.

A very effective valve saturation simulation is also incorporated, which adds those classic characteristics of warmth and fullness to the sound. The amount of drive is adjustable to allow the subtle valve sound character to be introduced — everything from clean to distinctly grubby!

The Auto-Tune facility is derived from the dedicated hardware and software processors, allowing you to create you own custom scales and to adjust the speed at which the pitch-correction process reacts. Another lozenge-shaped meter display (this time horizontal) indicates the amount of pitch correction performed and its direction (flat or sharp).

The Compressor/Gate section provides a variable-knee compressor with dedicated buttons to access directly the attack and release time-constants, and the slope of the knee. The threshold, ratio and gain make-up parameters are found by pressing the Comp button, and pressing this a second time recalls the detailed gain-reduction meter to the LCD window. The Expander/Gate section is accessed from a dedicated button which provides ratio and threshold controls. Another lozenge display indicates the total amount of gain reduction applied by both the compressor and expander/gate.

The de-esser module is equally predictable in terms of its facilities, with ratio and threshold parameters accessed through the De-ess button (along with the accurate gain-reduction display when pressed a second time). Two further buttons provide adjustment of attack and release time constants and the Highpass button adjusts a high-pass filter, setting the frequency region where the de-esser has most effect. Another metering lozenge indicates the amount of gain reduction being applied.

The final subsection provides a two-band equaliser, but this is actually a lot more flexible and useful than it sounds, because each band can be configured independently to operate in one of seven different filter topologies. These include high-pass and low-pass filters (6dB or 12dB per octave), high and low shelving filters (variable slope from 2dB to 12dB per octave), a peaking filter with ±18dB range and adjustable bandwidth from 0.1 to 4.0 octaves, and band-pass or notch filters. As a generalisation, and in keeping with many digital filter designs, the EQ section tends to suit corrective, surgical equalisation tasks somewhat better than creative ones. The parameters for each band are accessed by pressing the relevant Band button.

The On button in this section is divided in two, the top half engaging the equaliser and the bottom half selecting the double tracking facility. Pressing the Double Track button allows the type and amount of double tracking to be adjusted. The artificial output can either be dispatched as an independent signal via the dedicated secondary output at the rear of the machine (stereo mode) or it can be mixed in with the main signal to the main output (mono mix). In the case of the latter, the amount of double tracking signal can be adjusted from almost nothing to equal amplitude with the main signal.

The operation of the double tracking depends on the status of the Auto-Tune facility. If Auto-Tune is engaged to process the main signal, the double-tracked output is the original input. On the other hand, if Auto-Tune is not engaged in the main signal path, the double-tracked output is processed according to the Auto-Tune parameters (scale and speed). Careful setting of the Auto-Tune scale can therefore produce some interesting harmony!

The Output Gain button in this last section allows the output level to be adjusted (-30 to +24dB of gain), and there is another associated level meter lozenge. Pressing this button also provides access to the global bypass mode I mentioned earlier. If activated, the LCD screen continually shows a message that the machine is bypassed.

In the housekeeping menus are options to allocate the footswitch to control such functions as global or section bypasses — either singly or in combinations — and increment program presets. A MIDI dump mode allows the memory presets and configuration settings to be exported or imported as SysEx files. Memory presets can be handled individually or in subsets, and the machine configuration data can also be saved and restored independently.

In Use

The AVP1 is an interesting unit in a well-designed integrated package, and capable of some amazing effects. Many potential customers may confuse this machine with 'normal' voice channels, but this is so far removed from this familiar kind of mic preamp with EQ and compression that there really is no comparison. Antares are right to call it a Vocal Producer because that really is what it does best.

The mic modelling is fun, and can produce some very pleasing results from indifferent mics. It will not turn a poorly miked source into a good one, but it can help to impose some of the recognisable characteristics of other kinds of microphones. Close-mic recordings made in dead rooms provide the AVP1 with the best chance of delivering reasonably accurate results, but even if used simply as an effect, the modelling is useful. The tube saturation is something that many people will probably want to use all the time — that second-harmonic richness works wonders, particularly on male voices.

The Auto-Tune is fast and easy to use and can save the day with a vocalist (and some instrumentalists for that matter) who occasionally drifts out of tune. The handbook suggests setting the footswitch to bypass the processing until needed on the odd duff note — a handy trick which helps to minimise the processing side effects. And yes, it will also do the Cher warble if that is your thing...

The Compressor/Gate section does exactly what is says on the box and works very well indeed, with negligible overshoot when operating as a limiter, and very smooth action when used with gentle slopes and low thresholds. Being able to configure the gate as a gentle downward expander is also handy for controlling background noise levels in a fairly unobtrusive manner. Likewise, the de-esser worked well, although it would have been handy to be able to audition the side-chain, as this often makes tuning the filter considerably easier.

The Equaliser section is powerful, but seems to be a typical digital EQ design — technically correct but soulless. It is useful, but should be used with some care. Since the unit provides some unusual filters, such as band-pass and 24dB/octave low-pass, I had hoped to be able to create some interesting sweep filter effects controlled via MIDI from a sequencer. The problem is that the turnover frequencies are quantised to specific values, and so sweeps turn into lumpy steps which didn't suit what I was trying to do. Some may find this a useful effect, though.

I didn't really get on with the double-tracking effect either, but this may well come down to personal opinion again. There is a short delay to the double-tracked output, which helps, and there is obviously the retuning element courtesy of the Auto-Tune function, which can be quite interesting if only a few notes are permitted in the Auto-Tune scale. This facility could be useful in live stage work, but I fear it will have limited appeal in studio applications.

I was surprised that the I/O was entirely analogue, given that this is a digital processor and the preponderance of low-cost digital desks these days. I was even more surprised that the I/O was unbalanced, although, since the machine is intended to be used as a channel insert effect, and most mixers employ unbalanced inserts, this is unlikely to cause any problems in practice. Certainly, I didn't experience any interfacing problems in the studio and, overall, I found the AVP1 to be a really fun tool to use.

There is no doubt that this is ideally suited to stage use, and mainly with vocals, helping to create a more impressive and well-controlled sound from the typical stage dynamic microphone, while also tidying up any tuning inaccuracies. However, it would also be handy to tidy up fretless or upright basses and other freely pitched instruments. It is reasonably easy to set up and operate, and the provision of a number of preconfigured starting points helps a lot. It would have been nice if there had been a separate set of user program memories, rather than having to overwrite the factory presets, but this is only a small gripe.

Given that this is so well suited to live stage use, I feel it is a real shame that Antares didn't see fit to include a simple mic preamp and DI input as well. That would have made it an even more attractive and practical proposition for the live stage — but, again, I don't think that will cost the company many sales. Its facilities are sufficiently unusual and powerful that the lack of a mic preamp will not deter potential purchasers.

In the studio, when used as a tool during mixdowns, the AVP1 is a useful package with a good range of powerful facilities to sweeten vocal tracks and certain instrumentals. The inherent processing delay didn't cause any major problems during the review period, but should the processed and original signals become mixed there will be very obvious coloration and comb filtering. This happened on one occasion during the review when a vocal was being processed with a reverb plate program via a pre-fade send, and the feed to the AVP1 was taken from the channel's direct output. When the strong early reflections (with no pre-delay) from the reverb return mixed with the Antares output a nasty coloration resulted. Dialling in some pre-delay in the reverb cured the problem.

The AVP1 won't suit everyone, but if you feel hardware mic modelling and Auto-Tune would be useful functions to add to your studio, this is a very cost-effective way of proceeding — particularly if you want to be able to use the unit live on stage as well.

Pros

  • A generous collection of integrated signal processing.
  • The unique Auto‑Tune and mic modelling algorithms.
  • Valve emulation.

Cons

  • Wall-wart PSU and no power switch.
  • Unbalanced and all-analogue I/O.
  • EQ frequency parameter step intervals.

Summary

An intriguing combination of specialist signal processing optimised for vocal applications. Despite a few minor operational limitations this is an easy-to-use package which can help to polish a less than perfect vocal performance.

information

£568.99 including VAT.

Unity Audio +44 (0)1440 785843.

www.unityaudio.co.uk

www.antarestech.com