Page 2: Soma Laboratory The Pipe

The Algorithms

Orpheus features a pair of resonators with knobs controlling the size of each, plus their combined decay times. The pitch of your voice excites different frequency ranges of the resonators, while the Sensor adds extra bite or distortion. The results from this first algorithm are instantly impressive and include deep, ringing electronic drones, robot voices and mad pitch-bends. If you articulate with trills or sharp intakes of breath, the Pipe generates a hard, percussive resonance rather like striking sheet metal. I reckon that anyone with an aptitude for overtone singing might not turn the dial for a long time.

Filterra. If I told you this was a resonant filter with reverb, I'd be underselling it. The knobs control filter cutoff frequency, resonance and reverb depth/amount respectively, with the sensor acting as a freeze function. If you sing with freeze active, your unprocessed voice can join in with whatever is playing back from the freeze buffer. The manual describes a simple way to grab and hold a chord, for example, a trick that will be familiar to players of monophonic synths, ie. you drop an arpeggio into an ocean of reverb or delay so each note mingles with the others. When you release the freeze effect, the buffer decays naturally and you can, at that point, add more. This is a highly playable effect and one that responds well to hard edges, extreme tutting and the more percussive phonemes. When plunged into the deep and spacey reverb available from knob 3, my tuts generated some amazing cymbals, gongs and cavernous whooshes. Only the very top end of the filter's resonance gets a little too abrasive, at least for my tastes. If you draft in a looper pedal, the lush soundscapes served up by this algorithm can be used as a backdrop over which you can overdub some of the others.

Synth. Represented as a keyboard graphic, this effect features a low–pass filter and combined reverb/delay. A sub-octave is activated by the sensor.

The pitch follower works well enough for a selection of solo voices but I struggled to get pleasing results from this at first. This is possibly because I'd become used to speaking and singing quite loudly into the mouthpiece. Eventually I realised that subtlety is vital to avoid hard edges and distortion, although even that can be fruitful. For example, I managed to create a spooky but realistic impression of calls to prayer over overloaded speakers.

Singing softly with most of the mouth closed and the filter cutoff frequency low, pure tones are possible. Recently, I've been trying to simulate a Theremin but although I'm edging closer, my vibrato and control isn't there yet.

Reverb. Perhaps the best 'all-purpose' algorithm. This is a reverb plus distorted delay — the latter invoked by the sensor. You are given control of reverb time and mix, but not delay time; instead you can crank up the delay level and feedback simultaneously.

At the shortest reverb times, you can feed self-oscillating delays into small, reflective rooms then increase the size dynamically with the knobs. Similarly, tossing distorted clicks or complete washes of noise into a massive reverb soup will give soundscapes of the 'cinematic' or 'epic' variety.

Madelay. Like the sudden disappearance of a train driver at some remote station, this delay is strange and unsettling. It jumps around rhythmically through its data at a speed set by one of the knobs. Via the sensor you can freeze a small loop whose length is determined by another knob. The last knob sets the feedback amount.

It is indeed a fairly mad delay and a source of stuttering, repeating, clicking, ringing, droning effects. However, it's not quite so well-suited to regular delay duties, tape delay simulations and the like.

Pulse is an audio chopper with reverb. Control is given over the speed and width of the pulsing, plus reverb time. The sensor is used to restart the pulse stream, which can be useful when tweaking the speed to line up with an existing groove. Personally, I'd have preferred a sawtooth decay rather than the hard on/off that a pulse wave gives, and maybe a faster top speed would have been fun too. But if you're in need of gating effects or wish to simulate broken microphones, this algorithm is for you.

Bassdrum. The next three algorithms are drums and therefore ideal if you're in a beatboxing frame of mind. I never am, so all I can report is that they deliver thumpy kicks and snares with a fair degree of dynamic control. The knobs are employed to adjust the tuning, decay and pitch, plus the sensor either activates distortion or the snare.

Oktava features octave pitch-shifters, filter and delay. It's another valuable tool, especially if, like mine, your voice is a bit weedy. You can add octaves above and below your normal pitch, or switch to sub-octaves spaced at -12 and -24 semitones, the latter enough to give Darth Vader nightmares. A shimmering reverb/delay sweetens things up, although it's not really long or shimmery enough for my tastes. The filter is also quite subtle, which leaves an impression that this algorithm could have used more juice to really kick ass.

Generator is marked with a radioactivity symbol so you know things are getting serious. It consists of a weird, voice-driven oscillator, filters and ring modulation, plus a delay whose feedback varies according to input level. You do need to get quite worked up and shouty to set it ringing though, meaning it will suit some vocal styles better than others. The knobs control three frequency bands, plus delay.

The Pipe is also available in white. Both versions light up, so there's little chance of you losing it in the dark.

This algorithm is pretty wild and serves up bucket loads of harmonics ideal for impressions of throat singing, distorted ducks, didgeridoos, Daleks and every species of belligerent alien. External effects would really benefit this one as it can be quite dry and aggressive at times.

Harcho is, apparently, named after some kind of soup. I can only assume it's a soup full of rusty nails, razor blades and anchovies. If you're into cacophony this setting produces the nastiest, most distorted wailing I've heard in a long time. In theory you have control over the distortion mix, a filter and some reverb. Via the sensor you can switch into 'extreme' mode, which I would personally advise against if you want to remain on good terms with your partner.

I admit to hardly touching this one in the several months I've been tooting. Even reminding myself of it just now, I feel the need to go and lie down for a bit. Everything that it does sounds horrible. Some people will absolutely love it.

Conclusion

The human voice is our most instantly available means of expression. Via the Pipe, you can harness yours in new and unexpected ways — even if you don't sing particularly well. In this respect, Soma have created a genuinely new instrument whose appeal extends far beyond electronic, noise and drone acts, but which is clearly a natural fit for experimental or avant garde performers.

Each algorithm has a character that is only revealed after experimentation and practice. Therefore you won't get the best out of the Pipe without putting in the effort to develop a technique with the supplied microphone. While I expect that playing brass or woodwind instruments will help to a degree, it's open to anyone who can hum, sing, talk, beatbox or breathe.

I won't deny I found it heavy at first, but somehow that is in keeping with the Pipe's military-style construction and appearance. With a dozen different processing options available, some are bound to appeal more than others. For example, I'd definitely have voted to lose the drums in favour of alternate delay/reverb effects or maybe a softer synth voice or Theremin emulation. Similarly, I doubt I'll get much use out of Harcho or Pulse, but that still leaves more than half the algorithms to master, each practically an instrument in itself. I expect other users will have an entirely different list of favourites.

The Pipe is a very responsive instrument. From it I've coaxed droning choirs, edgy synthetic tones, distorted, lo-fi screams and eerie melodic material I'd never have attempted with my voice otherwise. Some users will doubtless focus on the distortion and percussion, while others work on impressions of tuvan resonant singing, didgeridoos and monochords. Whatever flavour appeals, this is a new and exciting reason to stand up and perform, either live or in the studio. Recently, in any recording situation where my inspiration has started to droop, all I've needed to do is reach for a Pipe and the ideas start flowing.