Theremin & Talking Machine

How about a rugby-playing countertenor?! Have you ever seen anything like THIS before? (I hope for his sake that this fellow is better on the pitch than he is on pitch).

http://www.youtube.com/watch?v=vn7gy-Yw_K0

Getting back to theremins and talking machines.

I have recorded a short piece with etherwave theremin, a ModTone BuzzBoy fuzzbox and an Electro Harmonix Talking Machine. As I have described elsewhere, the fuzzbox is in the effects loop of my (heavily modified) etherwave. The Talking Machine is in my effects chain after an echo pedal and is being controlled by an expression pedal.

I stress that this is not intended to sound like an opera singer. My criterion when choosing the fuzzbox as a way of adding harmonic content to the theremin sound was simply to find one I liked the sound of, not one that gave a plausible human voice, nor am I trying to play it in a way that resembles singing.

http://youtu.be/RO5jboNY3pk

You're getting a pretty awesome "burpy" sound on those low notes GordonC!

Are there many presets on the TM that you find useful, or is it mainly just the one?

I like that, Gordon!

Interesting to me also that the fuzz does not break through - the vocal filters are doing a damn good job! I strongly suspect that if you wanted it to sing opera, and you had Peter's playing skills, it would sound fine!

But anyway - I really liked the composition - and it went perfectly with the video.. This time the video was really needed to compliment the piece.

Fred.

Thank you, Fred. It seems that the more the fuzz, the sweeter the sound. I've got the distortion turned up full on the fuzzbox, and the lowpass on it set quite high so it's a fairly wasp-in-a-jar-ish sound before the Talking Machine gets it. (At least I think I have - if that is what the "fuzz" and "fizz" knobs do. :-)

Yup, gotta love the low notes, dewster - I think they sound like a bull bellowing lustfully rather than "burpy", but I guess there are plenty of interpretations (also I have never actually heard a bull bellow lustfully, and if I did I would definitely run rather than listen carefully!) I'm also enjoying the high pure resonant notes.

I think there's going to be two or three "go-to" voices that I'll use quite regularly and a couple of "special occasion" ones that I like but are so distinctive that it will be very easy to overuse them (one of them - the oy voice - I find so comical that I'll have to use it eventually, and then probably never again!) A nice feature is that you can program up your own presets, so if I wanted to use more than one setting in a piece I could step easily from one to the next by tapping the foot switch. 

I don't think I'll be using the built in LFO for a while - I like the control of having the filter sweep on the expression pedal. But it is possible to run the filter sweep on the LFO and the pedal together, and I have been experimenting recently with modulating volume with an LFO and the volume loop at the same time, and that can be quite cool. So that is something on my list of things to look at eventually. 

One very interesting feature is that you can plug in a control voltage source ins tread of the expression pedal. It expects 0-5 Volts, and the Etherwave Pro and Etherwave Plus supply 0-10 Volt control voltages, so it should be very easy to make a circuit to halve the voltage, and relate the filters in the pedal to either the volume or the pitch of the theremin. <rolls eyes> I was pretty sure that I wouldn't develop an interest in using control voltages, so I turned down the opportunity for Thierry to do the Plus upgrade while he had my etherwave, and now I've accidentally bought a device that seems to have a sensible use for them. Darn. Now he's going to look at me askance with a wry smile on his lips. But not for a long while, there is still plenty to explore before I get to that. And then Fred published a circuit a while ago that lets you scale and shift control voltages, for instance to turn 0-10 Volts into 2-4 volts, which could doubtless be augmented by circuits that lets you add two CVs in varying proportions, or turn 0-5 Volts into 5-0 to reverse the way the filters respond to changes in pitch, or volume, or both.

I also see from the documentation that it has a built in fuzz that has hidden controls - you have to push some button or other down until a light flashes or something. I haven't figured that out yet. Not sure if that comes before or after the formant filters. Either could be interesting - feeding the filters a doubly fuzzed signal, or fuzzing up a filtered signal. We shall see. I find it hard to predict the result of these things accurately - trying them out is necessary, and it's something I do very slowly. I like to thoroughly explore each setting before trying another one.

And then there is the way it interacts with my other pedals. I have a couple that are "special occasion" effects to the extent that I haven't found a use for them yet because I find the sound unpleasant (a treadle that ring-mods the signal against a sawtooth wave (I think), and a pitch shifter with a feedback loop) - I am definitely going to see what the Talking Machine does for them.

Already I have found it does wonders for the piece Void Ship, which uses a "special occasion" setting on my delay pedal - 5ms delay, max feedback, which turns it into a comb-filter, full of harmonically related resonant peaks. In the high frequencies the peaks are very close together and make a glissando sound like raking a key over the strings of a piano. But with the TM, when one of those peaks coincides with a resonant peak on the TM filters they are stretched out into icy shards of crystal resonance that make me go all poetic! And throughout the piece, it metaphorically turns all the knobs up to eleven. Maybe even twelve!

In summary, I think it's going to be a very useful addition to my toolkit but I don't think I'll be keeping the factory presets.

"I have never actually heard a bull bellow lustfully, and if I did I would definitely run rather than listen carefully!"

LOL - You should be ok in that situation, provided:

a) You dont look like a cow

b) Your not between the bull and a cow, or something that looks like a cow.

c) Bull doesnt think you are competing for the cow.

Your right - probably best to run!

;-)

"It seems that the more the fuzz, the sweeter the sound. I've got the distortion turned up full on the fuzzbox, and the lowpass on it set quite high so it's a fairly wasp-in-a-jar-ish sound before the Talking Machine gets it."

This is encouraging!  I wonder if I could feed the square waves from my digital Theremin prototype straight into it?  Or pulses, with the width modulated by the volume?  Two easy-to-do waveforms with nothing logic.

A couple of months ago (before I ran afoul of Theremin linearity) I was reading some papers and doing minor C simulation with vocal simulation, and was struck by the parallels between VOSIM, damped sinewave oscillators, and bandpass filter excitation by pulses.

I'm not sure, dewster. I tried passing the theremin signal through my Electro-Harmonix Micro Synthesizer set to all square wave and the Talking Machine didn't seem to do all that much to the sound compared to giving it the unadorned theremin signal. :-( I am suspicious of the Micro Synth though, I suspect it is doing more than just generating plain square waves.

I have an app on my iPod Touch (Bebot) that can generate a continuously variable frequency PWM signal, so I might try that when I get the opportunity.

In other news, I tried out my Snarling Dogs Mold Spore (combo wah-ring mod treadle) without bothering to defeat its noise gate with my effects loop and dialled up one of the zanier factory presets that uses the TM's internal LFO to sweep its filters and WOW! I know some very earnest people who spend literally pennies savaging flea market Casios to make sounds that are not half as whacked out and aggressive. 

Gordon, I love your Talking Machine video! It is entertaining and innovative. The challenge for the experimentalist and FXist will be to come up with new and different ways to use the sounds so that they don't become repetitive. 

"I wonder if I could feed the square waves from my digital Theremin prototype straight into it?  Or pulses, with the width modulated by the volume?" - Dewster

Hi Dewster -

IMO, the most important thing to think about WRT the input signal, is probably harmonic content. A square wave (50/50) only contains odd harmonics, and for this reason is extremely limited - Triangle waves also only contain odd harmonics (less intensity of harmonics than square wave) - one of the reasons I believe that theremins using XOR gate as the mixer sound 'dull' - ok for some sounds, but hopeless for strings or vocals.. In order to make square / triangle waves "interesting" distortion must be applied so as to add even harmonics (I believe this is done in the E-Pro)

If one produced a triangle wave, AC couple it, and chop off (rectify) one half cycle, one ends up with a waveform quite similar to what the larynx produces - Alternatively, if you adjust the width (M/S Ratio) of a square wave, and heep this ratio constant over the frequency spectrum (feed a constant amplitude triangle wave into a comparator for example) you can get a reasonable larynx simulation.

Modulating PW with something like volume could be interesting - but, in general, I have found that variable pulse widths can have dramatic and unexpected results when applied to formant filters - these could be useful if one wants 'experimental' sounds, and may be "good" when applied to the TM, but with my crude analogue formant filters the problems outweighed any advantages.

Another really good excitation signal for formant filters was a ramp waveform where partial (adjustable) 1/2 wave rectification was applied - this gave an extremely rich (often too rich) source of harmonics. An undistorted ramp waveform was a good source (much better than square or triangle) but was not as good as a 1/2 wave rectified triangle.

Fred.