TC-Helicon’s VoiceOne processor provides vocal modeling, pitch correction, and pitch-shifting functions in a rackmount unit.

In a perfect world, every vocalist would sing in tune and place each phrase neatly in the pocket. Capable harmony singers would hang out on every street corner, and no one would ever catch a cold. Matching the perfect singer to the track would be child's play, and the producer would never, ever decide to change the key after the singer has left for the Bahamas.

It's unlikely you live in a perfect world. Ever since the earliest days of recording, studio engineers have sweated to repitch warbling crooners, turn smooth-cheeked altar boys into would-be rock gods, tease a harmony track from thin air, or mask the nose job that a diva had between the original session and the overdubs.

Combining voice modeling (VM), intelligent pitch correction, and spot-on pitch-shifting, the VoiceOne from TC-Helicon puts the most-essential vocal manipulation tools into a single rackspace. Unlike the VoicePrismPlus, it offers no dynamics processing, EQ, or effects. Nor do you get four harmony voices, although with a little effort the VoiceOne can create stunningly realistic harmony parts that far surpass anything I've heard previously.


With an attractive metallic-blue faceplate, the VoiceOne shares the look of several of TC Electronic's popular effects boxes. Controls are logically arranged around the display. Utility functions nestle around the data wheel, and buttons for editing various VM and pitch parameters are grouped in the center (see Fig. 1). Two rotary pots offer manual adjustment of input and output gain over a 24 dB range relative to internally set levels. Dual ladder meters reveal input and output levels at a glance, and, as an unintended consequence, give visual confirmation of the processing delay.

Just as all controls and displays are on the front panel, all connections are on the rear (see Fig. 2). The most obvious difference between the VoiceOne and its predecessors is the lack of a microphone input. Analog ins and outs are on balanced XLR connectors only. Given that the unit accommodates a wide range of levels, I'm surprised that TC-Helicon didn't opt for Neutrik combo connectors, which offer ¼-inch and XLR inputs. Oddly, only one of the two analog inputs is active at a time, and the two outputs are split between the dry (unprocessed) and wet signals. To compensate for conversion and processing delay, the dry signal is delayed to match the wet signal.

Digital I/O is on RCA jacks, but the unit can switch from 24-bit S/PDIF to AES/EBU. All processing is at 24 bits, and you can dither the output down to 20, 16, or 8 bits. The VoiceOne supports 44.1 kHz and 48 kHz sampling rates with internal or external clock. Completing the rear view are the usual trio of MIDI jacks, a power-cord receptacle, and a ¼-inch TS jack for a footswitch.


The first thing you notice is the large display. Hats off to the design team, which managed to squeeze a lot of information into a limited space. In addition to the level meters, it provides information on the current preset or edit page, sampling rate, base key for pitch-shifting, MIDI status, pitch recognition status, and other functions relating to pitch correction (which I'll discuss later).

The display's right side is home to a row of six bar graphs showing the relative depth of each VM parameter. Corresponding buttons on the faceplate toggle each effect on and off. Double-clicking on a button conjures up the appropriate edit screen. You make edits using up and down arrow keys and the data wheel — a sometimes tedious process, considering the number of choices for some parameters.

Although the VoiceOne's user interface is consistent with those of other TC products, I found a few oddities. The only way to exit an edit screen is to choose another parameter to edit or press the Recall button. But be careful: pressing Recall twice erases your edits by recalling the previously saved preset. More confusing is that the VM depth meters are active even for parameters that are disabled. More than once, I had to confirm that an effect was on or off. Fortunately, there's a status light on each button.

Overall, editing on the VoiceOne isn't as easy as it should be, and the documentation is tough to understand. If TC-Helicon's writers can't provide an intelligible manual, they should at least proofread the one they have. And they shouldn't make users waste their time downloading names and descriptions of all the parameters from TC-Helicon's Web site (see the sidebar “Online Resources”).

A total of 99 factory presets showcase the VoiceOne's capabilities and serve as a starting point for tweaking. Edits can be stored to any of 50 user locations — an adequate number, but I wish I could overwrite the factory presets. Just how many times do you need to dial up something called Can't Sing, except, perhaps, to scare the pants off an arrogant client?


Although the VoicePrismPlus introduced voice modeling, the second-generation algorithms in the VoiceOne are a significant improvement, with greatly expanded control and realism. Just as digital modeling re-creates the sound of classic guitar amplifiers, VM simulates the characteristics that make a human voice unique: the length of the vocal tract, the shape of the nose, breathiness, vibrato, inflection, and even the rasp of overextended vocal chords. Because each singer is different, it isn't simply a matter of superimposing a model of a famous singer and, voilà, instant Bing. But used subtly, VM is a powerful tool.

Here's a rundown of the VM parameters and what they do. Inflection adds scooping (a singer's characteristic of starting above or below a note), portamento, and doubling effects. A variety of humanizing parameters control the inflection's onset, amount, randomness, pitch variation, and timing.

Thanks to an algorithm called Flextime, the processed vocal slows down or speeds up independently of the source. At times, the processed signal hits the note first! Obviously, that requires the dry signal to be delayed significantly; a little deft track sliding in my audio sequencer puts things back in the pocket. For greater realism, Time Randomization decouples the timing of the two voices. For doubling or harmony effects, the results are uncanny.

Vibrato brings pitch and amplitude modulation to the voice. Styles named Ballad, Broadway, Classic Rock, Nervous Tremolo, and Opera Tenor — 50 in all, including my favorite, Sheep — make it easy to find just the right vibe. I noticed a clear improvement over the first-generation VM effect: for the most part, the vibratos were rich and quite realistic. Controlling vibrato depth with my MIDI keyboard's modulation wheel was a hit-or-miss proposition. Although I liked the ability to choose the onset of vibrato, I often overshot the mark and added way too much.

Spectral styles refer to various EQ curves inherent within an individual voice. In conjunction with the resonance parameters, they create everything from subtle to drastic transformations. With 27 styles to choose from, the best way to hear how they work is to loop a vocal and scroll through them. If you need to sweeten a nasal vocal or add depth to a wimpy harmony part, Spectral styles are a good place to start. I'd be tempted to buy a VoiceOne simply for the improvement this parameter made on my voice.

Resonance styles change the voice's harmonic content in more dramatic ways. Some, such as Fat Tongue and Sumo, are fairly descriptive. Others, including the nine Transmute styles, have to be auditioned. If realism is your goal, a little of the Spectral style and Resonance style goes a long way, but there's some wild and crazy fun to be had here.

Two additional parameters, Breath and Growl, are self-explanatory. These parameters are best used in moderation, and a little goes a long way. Of the two, I think Breath might be the more realistic. A parameter called Breath HarmX simulates the harmonic shifts in a natural voice as the vocal folds open wider.

Taken together, the VM parameters are impressive. No matter whether your goal is to double a lead, create phantom harmony vocalists, or subtly alter the character of an existing vocal, you will find what you need.


Pitch correction and pitch-shifting are two sides of the same coin. Correction is used to fix individual notes in a slightly out-of-tune vocal; pitch-shifting generally implies either creating artificial harmonies or transposing an entire track. The VoiceOne handles each task remarkably well, though in slightly different ways.

As someone who has spent a lot of time recording vocalists who make up in enthusiasm what they lack in technique, I am a firm convert to automated pitch correction. Consequently, I was surprised to discover that my first attempts with the VoiceOne yielded less-than-stellar results. A quick trip back to the user manual revealed the reason: you must select not only a key and scale (as is common to both hardware and software pitch-correction processors), but also the window in which the effect will operate. In other words, if the voice wavers at 150 cents above the target but the window is set to 100 cents, there will be no correction.

Once I got the hang of it, though, I was sold. By setting the correction window suitably wide and manipulating the correction amount, I could fix large errors in pitch while letting the smaller deviations around the note remain, and the results sounded very natural. As a test, I corrected a vocal on the fly, recording the natural and processed vocals on adjacent tracks. Aside from the processed track being noticeably more in tune, the two sounded virtually identical. I heard none of the warbling and chirping that I have heard from other pitch-correction products.

For pitch reference, you select a key, then enter a scale or choose from one of 50 factory scales. Everything from standard diatonic, minor, and blues scales to various jazz, pentatonic, and non-Western scales are offered. The latter are somewhat problematic; the so-called Hawaiian scale is essentially a minor pentatonic variation, despite the overwhelmingly major orientation of postcontact Hawaiian music. Likewise, the Arab scale does not correspond to any of the maqam I could find in my library. My guess is that the designers simply totaled up all the permutations of five and seven notes available on the keyboard and doled out colorful names. It isn't a huge failing, but I'd love to see some real microtonal choices offered.

For easy reference, notes of the selected scale are shown on a tiny keyboard in the display. Hollow circles on the keyboard show the source vocal's pitch, while a pair of horizontal meters displays deviation from true (±200 cents) and the amount of correction applied. If you select MIDI in the Scale menu, correction occurs only when you hold down a note or group of notes.


The VoiceOne handles pitch-shifting tasks as well as any other processor I've heard, thanks to some very cool algorithms to discourage chipmunks and other common artifacts (see the sidebar “Avoiding Rodents”). To shift pitch as much as two octaves, select Chromatic mode and spin the data wheel. With accelerated data entry, it takes five revolutions of the wheel to twist your way from -2,400 up to 2,400 cents. Although you can't input numerical values directly, you can shift by scale degrees relative to the scale selected in the Correct Scale menus. However, this requires you to move back and forth between the Correct and Pitch edit windows.

As with pitch correction, you can control pitch-shifting with MIDI data, albeit on a separate channel. And because pitch-shifting overrides correction, you can accomplish both tasks simultaneously, correcting a vocal on the fly and force-shifting selected notes as you wish. That's right, you can replay a vocal line's melody on your keyboard. I can't wait to hear how some creative producer manages to abuse that amazing feature.

Although its primary mission is not harmony processing, the VoiceOne excels at the task. Simply choose a scale degree and select from one of three major and three minor scale-transposition maps, or create your own. Adjusting VM portamento parameters smooths the transitions between harmony notes. Additional adjustments to formant and Flextime parameters make harmonies sound eerily realistic. If I hadn't recorded the track myself, I would have sworn a second singer had entered the room.


Whereas the VoicePrism and VoicePrismPlus straddle the live performance and project studio markets, the VoiceOne is first and foremost a studio tool. Live performers might be put off by the processing delay. But latency can be diminished at a small loss in vocal quality.

The VoiceOne's voice modeling is a big improvement over the technology's previous incarnation, but it still has a long way to go. It's easy to overdo the processing, particularly with the Breath and Growl parameters. But with care, you can create stunningly realistic effects. The pitch correction and pitch-shifting functions are far ahead of anything else I've heard.

The VoiceOne is a great toolbox. If you work with vocals, you'll find a million reasons why you need one.

Acoustic musicianMark Nelsonproduces the Aloha Music Camp, a weeklong immersion into Hawaiian music and culture held in August on the Big Island of Hawaii (


vocal processor


PROS: Realistic pitch-shifting and pitch correction. Extensive control of vocal formants, inflection, and vibrato. Multiple pitch-shift and correction options.

CONS: No ¼-inch analog I/O. No direct input of numerical parameters. Poor documentation.

TC-Helicon/TC Electronic (distributor)
tel. (805) 373-1828

VoiceOne Specifications

Analog Audio Inputs(2) balanced XLRAnalog Audio Outputs(2) balanced XLRDigital Audio I/O(1) RCA in, (1) RCA out; 24-bit S/PDIF, AES/EBUA/D/A Conversion44.1, 48 kHz; 24-bit, 128× oversamplingMIDIIn, Out, ThruControl Jack¼" TS footpedalPresets(100) factory (1 blank); (50) userDisplay23-character, 280-icon STN-LCDDynamic Range100 dB, 20 Hz-20 kHzTotal Harmonic Distortion<-94 dB (0.002%) 1 kHz, +20 dBu outputFrequency Response20 Hz-20 kHz, +0/-0.5 dB @ 48 kHzPower100-240 VAC; <15WDimensions1U × 8.2" (D)Weight4.1 lb.


As anyone who's ever played around with the speed control on a tape recorder or turntable knows, changing a voice's pitch alters its formant characteristics radically. Nearly a half century ago, by multitracking his voice and speeding up the tape, producer David Seville created the Chipmunks, earning himself a string of hit records and immortality of sorts.

To avoid sounding like a chipmunk or giant when you do pitch-shifting, you must re-create the formant characteristics of the original voice — a task that requires a fair amount of DSP power and causes a substantial processing delay. To overcome the problem, the VoiceOne features two basic pitch-shifting modes. Normally, using pitch-shifting along with any of the VM parameters yields the most realistic effects. You can easily correct for formant alterations created by pitch shifts or use VM to simulate the sound of a second vocalist.

However, in a live situation where even a tiny delay won't cut it, choose PureShift mode. That allocates all the DSP processing power to pitch-shifting and correction and completely bypasses VM processing. The result is greatly reduced latency at the expense of a slight degradation in quality. You retain some control over formants by means of a single formant editor. Positive numbers yield progressively more masculine or mature sounds, and negative values sound, well, younger. I discovered I could mutate my yawp into a credible imitation of a little old lady singing the national anthem without too much effort.

PureShift offers one additional formant control: the oddly implemented Hybrid Shifter Ratio. According to the manual, it controls the formants during shifting. At a setting of 100 percent, formants are shifted along with the pitch, and 0 percent is full formant correction. To simplify things, an Auto setting is also available.


The 26-page manual and skimpy quick-start guide barely scratch the surface of what's inside the VoiceOne. To get the most from the unit, you'll need an Internet connection. TC-Helicon's Web site features ample resources, including white papers, application guides, FAQs, an online tutorial, and even audio files from the mind-boggling demo CD, complete with producer's notes. I'd also suggest taking a run through the online pitch-correction workshop.

To learn how the TC-Helicon products differ from other pitch-shifters, check out the white paper titled “Pitch Shifting and Voice Transformation Techniques,” by Patrick Bastien. Likewise, the company's Web site is the only place you'll find a complete listing and description of all the VM parameters.