Theremin And Vulcan Harp

Fred,

I wish I could tell you more about exactly how the Vulcan lyre works but the fact is, I don't know. It's like my car - I can drive the thing but I haven't the slightest idea how it works either!

I understand all the outside aspects of the lyre (buttons, knobs, tappers, strings, etc.) but the inside is a mystery. All I can tell you is that it is essential that the instrument be very exactly and precisely tuned before playing (I use an electronic tuner) otherwise things get "wonky".

"All I can tell you is that it is essential that the instrument be very exactly and precisely tuned before playing (I use an electronic tuner) otherwise things get "wonky"." - Peter

Thank you! The precision tuning (of, I assume, the strings, not the electronics) combined with the materials used, is a big clue as to its possible operation.

" It's like my car - I can drive the thing but I haven't the slightest idea how it works either!" - Yeah, im nearly in the same boat there ;-) - but with musical electronics I can (generally) understand workings better than I can actually "drive" the instrument.

If your harp does actually pick up both the "trigger" amplitude and frequency of each individual string, and resolves these independently into electronic signals, then the technology is beyond anything I have yet encountered - its not impossible that this is the case - I havent focussed a lot on stringed instruments because I find even a simple guitar impossible to play.. But I do not see any complexity at the bridge - which is where I would expect any transducers to be fitted - and with so many strings, I just cannot see each being picked up and seperately processed.. I can see the bridge acting as a composite pickup..

But whatever - I certainly couldnt "drive" the harp, and its quite likely that I dont have a clue about how it works! ;-)

Kind of makes me wonder whether theres some ET's working at MIT, and whether Star-Trek is a fantasy at all ;-) Perhaps its a genuine Vulcan harp, not the pathetic terran interpretation described in the star fleet manual. 

 http://cygnus-x1.net/links/lcars/vulcan-harp-construction-plans.php 

 ps.. Just realised another way it MIGHT work.. You say (or imply) that even minor tuning errors cause things to go "wonky".. This might be a big clue...

Perhaps the audio composite is filtered with an array of bandpass filters - sort of like a graphic frequency display - each filter tightly tuned to one specific string frequency.. From this, one could generate a voltage (or digital value) proportional to each strings vibrational amplitude, and would not need to derive the frequency as this would be known..

If the strings were off tune, this could cause havoc as the filters are fixed frequency, and all the data would be wrong..

Yeah - that all fits... One still has a composite acoustically generated signal to use (after bending if required) but one also has a set of values relating to each strings 'activation' which one could use to drive even quite a simple synthesis engine.

Funny thing here is that one could almost drive the simple top-octave generator as shown in the star fleet "Vulcan Harp construction plans" from the above - Ok, those parts are rarer than rocking horse poo these days - but a single FPGA and a few analogue bits and one could have a reasonable board for a basic Vulcan Harp...

Fred.

The more I think about the idea, the more interesting it becomes .. Its an idea which cannot work on a guitar or any other stringed instrument where the pitch of a string changes - but for a harp, its extremely elegant and simple.. So simple its beautiful! .. With any harp or instrument having the correct FIXED FREQUENCY tunings,one could implement this kind of detection - probably even by just using a microphone. If the instrument is acoustic and has a functional resonator, bending would not work - but all (or many of) the chording structures could work, and synthesis engines could be driven from the 'decoded' outputs... In fact, even MIDI could be output... And latency could be quite low - probably in the order of 1/2 cycle as no frequency tracking is required...

*I also wonder if there is any acoustic signal getting to the output - perhaps the synthesis engine just uses the string amplitudes to drive a high quality wavetable or similar engine that sounds like a string - this would remove the need for any pitch shifting of the composite acoustic signal.

------------

It might be quite simple to test the above hypothesis.. if repeatedly plucking a single in-tune string, one hears this being played through the audio output, and one slowly de-tunes this string.. If one is listening to the acoustic output, one should hear the pitch change.. If, however, one is hearing a synthesised output, the pitch should stay constant but the amplitude will probably change.. At some level of detuning, if the hypothesis is correct, the pitch (if one is hearing only a synthesised string) should jump to the nearest other string frequency - or one may hear two strings being played, probably at low level, and as the tuning gets closer to that of the other string, the amplitude should increase and the other strings frequency should play, even though the wrong string is being plucked...

If, when detuning a string, you dont hear this detuning coming from the speakers, then you can be damn sure that everything one hears from this harp is electronically synthesised... If you do hear the detuning, then one is hearing acoustic signal from the strings - and electronic synthesis signals from other voices. If all sounds one hears are synthesis, then my first hypothesis is entirely eliminated, and this latest hypothesis is possibly correct.

I know - it probably doesnt make any difference - but if you could observe whether you can hear the tuning through the speakers next time you need to tune the instrument, I would be real interested to know! ;-)

"It is a highly sophisticated synthesizer that uses the strings of the instrument as triggers"

Yes - I think this is the truth - my mis-thinking was probably with regard to how the triggers from each string were derived, and seeing the strings were tuned, the assumption that pitch tracking was required.. Closely listening to the plucked string sounds I am now feeling these are probably synthesised and not acoustic - Its difficult to tell because you play so well - but the sounds might be just a little too perfect, even for you! ;-) - and the bending is perhaps a little too clean and quick to be done by DSP.

 It would be exciting to discover that I was wrong with regard to how individual strings were resolved, as this would possibly mean that id thought of the idea first, LOL ;-) But im reasonably sure now that the Vulcans invented it, and am quite satisfied if I have merely worked out what they did !   ;-)

I just watched the video again.  Peter, please tell me if I've got any/all of this right/wrong:

1. There are two components to the sound, the direct string sound and some kind of "harmonic wash" in the background. 

2. The "valve" buttons control the "wash" but have no control over the direct string sound.

3. The electronic pitch bend knob only applies to the "wash" and not to the direct string sound.

If the above is correct than it's just a harp with adjunct harmonic synth.  Though a very interesting one!

"If the above is correct than it's just a harp with adjunct harmonic synth.  Though a very interesting one!" - Dewster

It would be really great if one of the inventors of this instrument paid us a visit! ;-)

To me, the "adjunct harmonic synth" idea does not logically fit the data given regarding the Lyre's operation - If it was just a "adjunct harmonic synth" then the tuning of the strings would have no effect on the operation of any synthesis engine.. Peters statement ""All I can tell you is that it is essential that the instrument be very exactly and precisely tuned before playing (I use an electronic tuner) otherwise things get "wonky"." Is what completely removed any doubt in my mind over any question that the strings were unrelated to the synthesis.

Everything I have suggested is purely speculative and hypothetical, and I try to test these against any new data given.. I still have a few areas of uncertainty, your last post neatly encompasses all these areas:

1. There are two components to the sound, the direct string sound and some kind of "harmonic wash" in the background. 

I am not sure whether the "direct string sound" is from an acoustic "pickup" of the string signals or whether its electronic synthesis.

2. The "valve" buttons control the "wash" but have no control over the direct string sound.

This is an interesting issue - I am inclined to think that the "valves" do not affect the "direct string sound" (however thats produced) but that the strings actuated do affect the sounds produced by the "adjunct harmonic synth"

3. The electronic pitch bend knob only applies to the "wash" and not to the direct string sound.

It was my near certainty that the "direct string sound" is affected (bent) by the knob which caused me to propose my original hypothesis, which was that the whole instrument was a complex DSP / pitch shifter using composite acoustic signal as a source.

The fact that string tuning is critical for non-wonky operation has almost completely undermined that hypothesis though.. Which leaves a possibility of real-time pitch shifting on an acoustic composite, combined with pitch bending on the synthesised voice (getting both to track could probably be a difficult complex task) or that everything heard is synthesised (a lot easier to implement).

You have presented another option - that the direct string sound is acoustic and not bent... This idea conforms closely to the original documentation, But I am inclined to think this Lyre is more advanced - I believe that I can hear the "direct string sound" (the decay of the bass string in particular) being bent on several occasions.. I also strongly suspect that the "adjunct" is far more than could be obtained exclusively by use of the valves - that the strings are providing data to whatever engine is providing this 'adjunct'.

On all the above, however, I have a high chance of being wrong - I think it would probably be a lot easier to work out whats really going on if some incompetent was playing the instrument - The trouble (from a technical perspective) is that the performer is a master musician - and "trickery" (even non-intentional) is a part of the whole presentation.. By selecting the right strings and buttons, and perhaps 'twanging' a different string slightly while performing a bend, he could quite easily probably convey an illusion of something that isnt "really" happening.. Even the "Adjunct harmonic synth" MIGHT just sound to me like the strings are influencing its output because the musician selects the chord or whatever and matches the strings played to this...

The other "problem" is that the musician and the engineers may have only a fairly limited common vocabulary with which to discuss technical aspects - And when my car needs repair, I have limited interest in what the mechanic has to say about the finer points - I will politely listen and nod as the virtues of some topology are expounded over another, but couldnt say anything useful even if I was interested! - Even worse - I could ignorantly mislead the mechanic by saying the car was blowing grey smoke, when he would describe it as black smoke..

Fred.

ps.. Dewster, were it not for the "the instrument be very exactly and precisely tuned before playing (I use an electronic tuner) otherwise things get "wonky"" I would probably be inclined to thinking you may be right.. It is this one disclosure upon which my entire hypothesis hangs... But hey, that may be "grey smoke" vs "black smoke".

Hmm, there does seem to be very definite pitch bend of the strings as well as the harmonic wash.  The string sound is pretty realistic, I'm inclined to think it isn't synthesized.

I wonder how all those controls festooning the back work.

"I wonder how all those controls festooning the back work." - Dewster

LOL ;-) .. Im not even tempted to venture a guess ;-) ... That would be a bit like trying to speculate on the origin of the universe when one hasnt even verified that the earth isnt flat!

The only information we have is that these controls are "locked" during playing - I think one can guess from that that they are scanned and probably digitised.

But whatever - visually the whole instrument is a masterpiece - even if those controls did nothing, they need to be there in that form IMO - It looks like its from another world! - The fact that its also a superb instrument and that the controls actually do something (whatever that may be) - well - I would really love to meet the person/s who designed and/or crafted this wonder!

Fred.

"The string sound is pretty realistic, I'm inclined to think it isn't synthesized."

So was I - the strings do sound really good... But even here theres a conflict, because we know they arent physically detuned by the bender (as you observed), so this means that if one is hearing the strings, they are being processed by a RT pitch bender.. and I have never heard any RT pitch bending which sound this good - particularly on string sounds or sounds where amplitude is changing quite quickly and pitch is being bent quite rapidly..

So we either have extremely well synthesised strings being bent through a simple pitch bend control which also drives the 'other' synthesiser voices, or we have an acoustic string sound being bent through an extremely high quality RT digital processor, and the bend control also being appiled to the synth engine.. 

Neither of the above fit well with what my ears are saying - but unless I am missing something, it must be one of those two options. I have heard some really good synthesised strings, particularly plucked - Where these can often be identified as different from real strings is down to unintentional sounds and resonances / distortions etc which always happen to some extent with real strings - scratching, buzz, sympathetic resonances - that sort of stuff which, even when well below 'noticable' level do give a real string qualities that cannot be replicated easily with a synthesised string - As I said before, I think that PERHAPS the string sounds from this lyre are a little too perfect - but that may just be because the musicians playing is close to perfect!

Also, I am well out of date on what is possible using DSP on audio - nothing I have is less than 10 years old, and apart from my MPX-1 no digital processor that I own is particularly outstanding - So it may well be that bending has advanced since I last looked into it! ;-)

Note that there are videos of another instument (http://www.youtube.com/watch?v=gN9yUUd2_J8&feature=related) - one of several - made by a luthier that is, at least visually, more consistant with the lyre shown in Star Trek.  I didn't see any videos of anyone actually playing it though.

Those were made by Gadson Huckabay.

From Gaston in an old email...

The 2 dials you're referring to are what I call "voice dials" and yes, each work independently of each other.  The bottom controls depth modulation and the top handles rate modulation."

https://s.amsu.ng/XVkvA6wwI9sN

More pics

https://s.amsu.ng/0V3wK23n7mmN