You are here

TC-Helicon Voice Works

Mic Preamp & Effects Processor
Published June 2003
By Paul White

TC-Helicon's new rack unit aims to provide you with every conceivable vocal recording tool, including EQ, compression, pitch-correction, double-tracking, intelligent harmony generation and effects.

TC-Helicon Voice Works header photo.Photo: Mark EwingDesigned for use both live and in the studio, the 1U TC-Helicon Voice Works brings together the processing expertise of TC Electronic and the vocal processing technology of TC-Helicon to combine a fully featured voice channel (incorporating phantom powered mic preamp, compression and EQ) with four-part harmony generation, vocal doubling and vocal pitch correction. To add polish to the finished result there's a reverb section and a tap tempo delay, and the unit also features full MIDI control. Everything other than the mic preamp is digital, operating at a resolution of 24 bits with either 44.1kHz or 48kHz sample rates.

The user interface is reassuringly straightforward, with dedicated buttons accessing all of the key features and menus directly, and, for use post-recording, there is a line input to allow existing vocal recordings to be processed. The analogue mono line input and stereo line outputs are on balanced jacks, with further S/PDIF I/O on coaxial connectors. The mic input is a conventional balanced XLR, and a full set of MIDI In, Out and Thru sockets are fitted.

TC-Helicon Voice Works £623
pros
  • Useful combination of voice channel, effects, vocal pitch-correction and harmony generation.
  • Friendly user interface where presets are used intelligently to make operation easier.
  • Quality TC Electronic effects
cons
  • The harmony voices still sound artificial when subjected to scrutiny, though they are an improvement over anything that's gone before.
summary
Voice Works combines several key TC Electronic and TC-Helicon technologies in a single, affordable unit without overburdening the user with complexity.

sos_wob.gif

The key to making a product of this kind sound good is realistic harmony creation, and with every generation the TC-Helicon technology gets better. Voice Works employs something its designers term Hybrid Shift harmony generation, which was developed for the TC-Helicon Voice One. Hybrid Shift is described as using a dynamic combination of different pitch-shifting technologies, and is designed to produce a more natural sound than traditional formant-preserving methods, with fewer audible artifacts.

Four-voice Harmonies

Voice Works creates up to four simultaneous voices of harmony in four different modes: Scalic, Chordal, Shift and MIDI Notes. Scalic mode follow a preset or user scale and key, while Chordal mode utilises an internal chord recognition algorithm that automatically creates harmonies that fit the chord sequence. MIDI Notes mode allows direct control of the harmony notes via MIDI. Amongst the tricks used to make the artificial harmonies sound more realistic are Flextime (to simulate the timing discrepancies of real performers), vibrato, inflection, pitch randomisation and portamento. These, along with clever algorithms that smooth out pitch changes when moving from one scale note to the next help break up the mechanical precision of the process and give each of the harmony voices a slightly different performance character. It's also possible to shift the formant of each voice to make it sound more male or more female, and the range is sufficient to go beyond the boundaries of the natural into the realm of special effects.

A newly introduced Harmony Hold feature allows the currently sounding harmony to be held indefinitely using a sustain-style footswitch and, where harmonies are not required, the processing can be used to simulate layered voices or double-tracking effects. Because the Voice Works is designed both for live and studio use, it has the ability to store data for 50 Songs. Each Song can contain up to 30 Presets to allow for things like key changes or choruses that require a different scale structure to the verses. Note that trying to perform live with vocal pitch-correction being applied is rather difficult, so it's usually best to arrange for an uncorrected version of the vocals to be fed through the stage monitors.

The vocal pitch-correction is essentially lifted from the original TC Electronic Intonator and, again, it is scale-based, forcing off-key notes to the nearest valid scale note. Various controls are provided to keep the sound natural, including rate of correction and a facility for progressively bringing the pitch closer to being in tune without forcing it to conform to the target pitch exactly. A windowing facility allows the user to set limits, so that notes outside the pitch range of the detection window are left untreated.

User Interface

One welcome feature of the Voice Works is its intuitive user interface. The line input gain control is on the extreme left, followed by a parameter window that normally displays the intervals to which the harmony voices are set, as well as patch, key and edit parameter information. The mic amp gain control is located at the opposite end of the front panel and up to 65dB is available.

A Preset comprises eight components, directly addressed via the four voice keys and the Harmony, Thicken, Effects and Pitch Correct keys below them. The Effects key also doubles as a Tap Tempo button for use with delay effects. Pressing a key toggles between that function being on or off, whereas double-clicking the key accesses the associated edit menu. This makes it very easy to add or remove effects or to experiment with the number of harmony voices.

 Test Spec 
 
  • TC-Helicon Voice Works OS v1.04
 

Song mode enables the user to step through the Presets that are used within a Song. In live use, this can also be accomplished via MIDI, a simple footswitch, or a triple footswitch. The Setup key accesses globally defined functions (independent of the Presets) where the EQ, compressor/gate, digital audio settings and MIDI configurations can be adjusted. Pressing the Setup key moves between the main setup functions, while the arrow keys are used to access individual functions within that group. Pressing and holding an arrow key accelerates the rate at which the parameters are accessed. Recall is used to recall Presets, while Store is used to save the current edit state as a new Preset. Parameter value changes are made using the data wheel to the left of the Enter and Bypass buttons in the Control section. The small Mic In button switches from mic to line input — in mic mode, the line input may be used as an auxiliary input. When used as an aux input, the line input bypasses the harmony and pitch-correction sections, but is able to use the effects if required. Two further buttons activate the phantom power and introduce a 20dB pad.

Using Voice Works

Before starting to program any Presets or Songs, the Setup section needs to be configured by the user, as it's here that things like compression, EQ, lead voice input level and pan are set up. Dynamics processing can be applied to the lead voice, the harmony voices, both or neither. The EQ section offers three processing bands, with adjustable high and low shelving sections and a fully parametric mid-band. Again this may be applied in any permutation to the lead and harmony voices. Another neat humanising feature is Lead Voice Delay Compensation, which delays the lead voice slightly so that the harmony voices come randomly just before or just after the lead voice. Without any delay added, the latency through the system is only around 6ms, which is subjectively undetectable. I/O and sample rate settings are also made in this section.

Once the global settings have been made, Presets are created in much the same way as on a synthesizer, insomuch as the current Preset is held in a memory buffer and may be edited, but no changes are permanent until the Preset has been saved. A Song comprises one or more Presets and, though you can step through Presets using a single footswitch, the optional three-button footswitch is better, as this also allows the user to step backwards through the Presets and to mute the harmonies where required. Up to 30 Presets can be used in one of up to 50 Songs. Preset names are appended by a letter 'C' to show they are chordal or 'S' to show they are scale-based. All Song changes are saved as they are made, so there's no formal store procedure to go through and creating a Song really is just a matter of chaining up the desired Preset numbers. Note that, where the triple footswitch is used, the footswitch functions are different depending on whether the Voice Works is in Song mode or Recall mode. When controlling the harmonies via MIDI, it is also possible to use the modulation and pitch-bend wheels on a typical MIDI keyboard to influence detuning and vibrato depth.

The standard automatic harmony generation relies on the user telling the Voice Works what key the song is in and then deciding which of the harmony voices to use. In the case of scale-based harmonies, the user can also create custom scales. It is further possible to choose between Equal Tempered, Just or Barbershop tuning, the latter using the input notes themselves as tuning references rather than always working from the root note of the scale. This is designed to work best with a cappella vocals to give the most true-sounding harmony.

Parallel pitch-shifting is also facilitated, though this is of limited use and is usually restricted to fourths, fifths and octaves. In Chordal mode, Voice Works uses the root, third, fifth and occasionally seventh notes of the chord, and sets the harmony voice to the closest note within the chord. However, in Scalic mode it simply pitches the harmony voice to the nearest note in the selected diatonic scale. A smoothing parameter decides how much of the lead voice's pitch fluctuation is imposed on the harmony voice, while the humanisation options (based on Flextime timing randomisation, pitch randomisation and scoop) can be chosen from a list of style types. It is also possible to adjust the attack and release times for the harmony voices. The organisation of the vocal and vibrato styles into simple preset memories, rather than letting the user loose on the individual parameters, makes setting up much easier, without sacrificing much in the way of tweakability.

There are two processing blocks applied to the lead voice: pitch-correction and lead voice thickening. Pitch-correction, when used, is applied before harmony generation. The user needs to enter the root note and scale type, then make any necessary adjustments to the correction window (outside which no correction takes place), the rate of pitch-correction (slower is more natural) and the amount of correction applied. The amount of pitch correction currently being applied is also shown in the parameter window, where a meter moves left or right of centre to indicate flattening and sharpening. Lead voice thickening creates two pitch-detuned versions of the input voice, where the user can control the amount of detuning and the stereo spread of the doubled voices to instantly create the illusion of doubled tracking.

That leaves the effects section, which comprises reverb and delay. There are 15 environment simulations ranging from Living Room to Cathedral, and these can be called up and adjusted for reverb time and HF coloration. The delays come as dual-mono or two variants of ping-pong — the time can be set manually, can be entered using the Tap Tempo button, or can be sync'ed to MIDI Clock. Variable high-frequency damping can be introduced into the delay feedback path and effects can either be saved per Preset or globally. The effects can be switched on/off separately for each of the lead voice, harmony voices and aux input.

TC-Helicon Voice Works rear panel.Photo: Mark Ewing

On The Session

As a mic preamp, the Voice Works is clean, and the addition of digital EQ and dynamics control gives it complete voice-channel functionality, though I'd always recommend recording without processing where possible, then using the processing on mixdown, as this keeps your options open. Given TC Electronic's track record with processing, it's no surprise that these sections work musically and simply, as indeed does the reverb, despite the minimal user adjustment. The vocal pitch-correction also works smoothly, provided that you're not tempted to make it too assertive, and it has a useful display of the amount of retuning being applied. If the correction rate is set too fast, the voice takes on an unnatural, slightly robotic quality, but at sensible settings and with a singer who's not too far from the mark, it's very natural-sounding.

That leaves the harmony-generation section, which tracks the lead vocal line and reads its pitch extraordinarily well. How natural the harmonies sound depends on the characteristics you choose for the various voices, though it's probably still fair to say that none of them stand up to close individual scrutiny due to the pitch-shifted character of the sound. In this respect, the shifted quality is far better than I've heard before, but if listened to in isolation then it's quite evidently not real, even though it can work extremely well within a mix where it's layered with a main vocal part. Of course you have to take care how you set up the harmony voices — very low gender settings in particular always sound to me as though the virtual singers are trying to clear their throats — the voices take on a growling characteristic. However, if the gender isn't shifted too far from the original, the result is far more convincing.

The ability to apply presets for vibrato type and vocal style helps get good results quickly, and of course the harmony voices need to be mixed appropriately with the lead voice to get a natural-sounding result. In many cases, using fewer than the maximum number of harmony voices sounds the most natural, as a very dense harmony can sound a little like a vocoder in some situations. Although I wouldn't normally use any form of artificial harmony generation in a mix where the vocals are very exposed, the system used here is as good as you'll find anywhere. In addition to its obvious live applications, it can be used to good effect in the context of many pop music styles. Vocal thickening is also handled effectively, giving a plausible double-tracking simulation, something that can be further enhanced by the careful use of short-duration delays.

Master Work?

Voice Works gives the user a very proficient voice channel combined with TC-Helicon's latest harmony generation and voice processing technology and one of the best vocal pitch-correctors on the market. Although the effects are simple and have little user editability, they actually sound very sweet. This combination clearly has many live sound applications, not least for the solo artist attempting the Beach Boys' greatest hits or Queen's 'Bohemian Rhapsody', but Voice Works also has a number of studio uses. It can be used both when recording and mixing, but, as I pointed out earlier, saving the processing until the mixing stage gives you the maximum flexibility.

As a voice channel, the Voice Works sounds clean and positive, while the pitch-correction is easy to set up and generally very natural-sounding. Double-tracking is a doddle, and provided that you are prepared to spend just a little time adjusting the way the harmony voices respond, it's also possible to generate surprisingly realistic harmonies. If you don't pay attention to choosing suitable harmony voice characteristics, or if you get too extreme with the settings, you can end up with something quite electronic sounding — which may be equally valid on an artistic level depending on the type of music you're working on.

Another application that's often overlooked is simply to use the Voice Works to create a choice of harmonies that can be learned and then sung by a real singer. The algorithms used to create the harmony lines are extremely effective, and Voice Works might just come up with a cracking good harmony line that you or your client hadn't thought of.

If the idea of having all these tools — preamp, pitch-correction, harmony processing and effects — in a single box appeals to you, then the Voice Works does as good a job as anything else you can buy, with the absolute minimum of complexity and at a very attractive UK price.

£623 inc VAT.

TC Electronic +44 (0)800 917 8926

www.tc-helicon.com

Published June 2003