Theremin Sound & Acoustics

- Thanks for the feedback, Italian, Fred and Chobbs -    developing webpage

Edit: My acoustic methods were done years ago before I was aware of background noise, if you can ignore that then you might enjoy a single Theremin in a Bucket dueling as a Theremin & Cello. This is real time register shifting.

I am not a musician but I think sound mods might be divided into two groups, distortion & enhancement. I mentioned to Italian _Thereminist about using an acoustic approach to get closer to the Cello sound he desires

This subject of a Cello sound comes up periodically and I thought this would be a nice project to explore for a couple of weeks.The past year has been a theremin marathon for me; my feet bleed and yet the adventure is about the fun of discovery rather than crossing a finish line. LOL

Theremin fundamentals are part of Nature; this theremin sound may have acoustically capture a moment in nature when suffering was allowed. The next morning the sun rose and all that lives celebrated.

A Cello is nice acoustically but let us not set limits on what a theremin sound should be. My theremin sound currently is like a single violin string stretched on a board. Acoustics is something I have avoided by using direct to sound card recording which is better for Analyzes of the electronic circuit behavior.

Many use the speaker to mic approach when recording to give the theremin sound a little more development.

I find electronic enhancement to be much "easier to control" where by using acoustics you are constrained in how nature wants to behave. It can be very frustrating.

My first thoughts on an natural acoustic enhancer experiment, it should cost less than $100, portable and self contained. Possibly develop it within the Diamond Speaker approach or in a small suit case (luggage). This is only practical if the everyday DIY individual can duplicate it.

I would not want this as my main theremin sound but something that could be faded into. In a way this gets back to what I call the third dimension of theremin. A two trick pony?

The electronic components I ordered for the experiment:

Amp/Speaker transducer & Piezo Pickup (EQ/PreAmp optional)

If there is magic it will be found in the method of acoustic chamber design which could be as simple as two pieces of 2 x 2 foot pieces of birch plywood paralleled, the thinner the better. 1/8" to 1/4"

I will report back on this idea in a couple of weeks if not scrapped. As every designer knows the theremin has more reasons not to work or to not sound musical!

Christopher

.

Dr. Maurice Martenot, inventor of the ondes martenot (the "sister" heterodyne instrument of the theremin) experimented extensively with various sorts of acoustic enhancement devices that could "shape" the sound of the ondes as it came out of its speaker.

The most famous of these was probably the "palme" which consisted of a large wooden palm shaped resonator, similar to a psaltery, with 12 chromatically tuned metal strings that would vibrate sympathetically with the sound of the ondes. This not only gave a warmer, more acoustic sound to the instrument but it also looked good on stage. Without it, the ondes is a rather uninteresting, small, celeste-type of keyboard with nothing particularly distinctive about it.

Martenot also used large gong-type resonators (le diffuseur métallique) that would provide a hollow reverberation. He experimented with a number of other devices, but only these two exist today. Here are a couple of pix showing a "palme" and a "diffuseur" for those who may be unfamiliar with them.

http://www.johncoulthart.com/feuilleton/2006/10/23/the-ondes-martenot/ 

Thank you Christoper for launching this topic!

I totally agree with you that electronic enhancements are easier to control... but please think about an average guy like me that has not much DIY skills!!!

Yesterday I was talking with my father and he told me that in the 60's and 70's was quite common to have very "analog" systems to produce effects... like putting speakers close to a vent or by quickly rotating them etc... I wonder if he knows any trick for a small acoustic chamber like the one you're thinking about.

Please let me know if you find something interesting in your research... I'll update the topic if my father has any idea!

Have a nice evening!

P.s: I've updated my profile image with my Kees Enkelaar theremin.

-The nice thing about creating webpages either private or public is it documents ideas as they pass by over time. I rarely recognize the significance of many things in the present, only in hindsight

Fred uses the words vocal formant which is somewhat new to my understanding but looking at the original sound byte in Audacity compared to this one below I think it also has them.

Cat & the Fiddle  mp3.    Homemade Cello Sound   (Fred's Favorite)

If I can reproduce the sound I will post how I did it, which is what the experiment is about.  If this second sound byte Cat & Fiddle has the character it is not done with a rubber bucket, rather a pure square wave driving a 12” wood panel. This is the direction the experiment seems to be taking us for now. Buckets have resonant points that are difficult to tame. I know the lower sound is a square wave because I was dividing down the fundamental frequency with a 4013 digital chip to get the register shifting effect.

With a little bit of luck just over-driving the transducer amp will square up the signal and allow the EWS to create a pleasant sound. I will allow the experiment to take me where it wants to go, not to necessary mimic any preconceived notions of theremin sound.

The greatest difficulty I have in experiments is maintaining a clean table top area to work on. (-'

Christopher

Edit: Fred has good ears, this brings me back to the rubber bucket which is actually an easier setup. With theremins nothing is easy!

"Cat & the Fiddle  mp3.

If I can reproduce the sound I will post how I did it, which is what the experiment is about.  If this second sound byte has the character it is not done with a rubber bucket, rather a pure square wave driving a 12” wood panel." - Christopher

This does sound much more like a fiddle / Cello to me! - And yes, you can hear the woods resonance! - One can hear some 'vocal' qualities, but they are IMO extremely weak compared to your bucket sample on the other thread.. One can hear these same 'vocal' qualities from many acoustic instruments.. and from all sorts of resonators.. But no resonator I have EVER heard comes close to the vocal qualities heard in the sample on the other thread!

Vocal formants are not "just" frequencies - one can have sounds with harmonics at normal formant positions, and these sounds may not sound like a voice.. You need the first and 2nd formant to start defining the vocal qualities (ooh, aah, ee or whatever) and the simplest formant filter may only peak at these - in order to change from say aah to ee both formant frequencies must change - and they are independent of each other - as in, they are not integer mutiples locked to each other.

One can create vocal-like sounds by using quite simple filters (the wah wah pedal is a common formant type filter)

But, in my expierience, to get the vocal qualities you presented on the other thread, you need at least 4 formants - and you would need the excitation signal to be perfectly matched.. I have only really played with analogue bandpass formant filters (these are formed by chaining a series of 1st and 2nd order highpass and lowpass filters one after the other, each of which must be tuned, and the level of each adjusted) - and this just gives you one fixed *formant* (composite set of independant formant frequencies forming the vowel).. To get changing *formants* you must adjust every formant independently (or have a processor with a table or whatever to adjust them) .. Digital formant filters as implemented in the TM are far more precise and 'human' - and the frequencies are probably not fixed hard as they are with a hard-wired set of bandpass filters, but move a bit dependent on frequency and amplitude of the incoming signal..

Somehow, this seems to have been acheved by your rubber bucket!

getting an occasional 'touch' of ooh or aah or other vocal 'effects' is not uncommon from electronically produced sounds - but these are usually brief, imperfect, and frequency dependent - one can sometimes splice a few cycles of some sound, chain these, and produce a "vocal" sound which one doesnt hear at all from the original.

Using acoustic resonators can greatly enhance the sound of electronic instruments IMO, particularly if you want to emulate an acoustic instrument.. With theremins, I have experimented a lot with making the enclosure vibrate - in 2010 I had theremins built into loudspeaker boxes, and they had an acoustic quality I believed came from FM capacitive feedback caused by the antenna vibrating..

I have used these transduces (well, not bought from Sparkfun, but these look identical) both as drivers and 'microphones'. https://www.sparkfun.com/products/10917

Fred.

 ps.. dont worry ;-) I have been delayed, but next week I will be gone! ;-)

Hi all,

Still working on that poor mans acoustic sound < $50. I have found a workable approach that prevents a lot of the issues when using acoustics. This is done with the stock EWS for the audio. The webpage for this is developing.

What do you think, opinions from comedians also welcomed. (-‘

EWS Theremin Acoustics  500k

Not the sound Fred likes but this can be improved upon. Waiting on Amazon to deliver more stuff.

Christopher

PS: Curly top: That was like graffiti on a mural, be careful.

"Not the sound Fred likes but this can be improved upon." - Christopher

Oh - I never said I didnt like the other acoustic modified sounds.. I just said that, from the perspective of vocal qualities, they didnt come close to the (IMO) exceptional sound in that rubber-bucket sample.

To me, any sound where the waveshape / harmonics do not alter as a function of pitch and/or volume, is uninteresting and / or irritating.. Such sounds, when "processed" through any natural / acoustic resonator, to my ears, always "improve"..

IMO, if one wants a Cello sound, the rubber bucket missed by a long way - BUT, IMO, it gave an astounding vocal 'simulation' - so astounding that I was really interested in knowing how exactly it was produced - NOT because I want to copy it, but because I am interested in the physics / mechanics involved.

The other samples you presented had interesting qualities, but were not any mystery to me..

When I say "NOT because I want to copy it" this is perhaps not entirely truthful - If I could derive a set of functions relating to how a rubber bucket, when 'excited', produced moving formants, it MAY lead to a simpler novel route by which I could produce a similar vocal formant synthesiser using electronics.... This is just the way my mind works - try to look at things, understand their operation, then apply any new understanding or 'grasp' to some other problem or implementation...

But the above process must start with valid data - the more data one has, the easier it is to hypothesise mechanisms to produce this data, or to decide whether further investigation is really worth the effort..... There are so many things to explore, and so little time!

I do think that the experiments you are doing in this area are really exciting and have great potential - so much so that I am "forced" to return to my domestic war zone to access TW briefly (I am in the process of moving, and wont have telephone or internet until mid October)..

I keep hoping that you will reveal more details about the rubber-bucket-resonator so that I can ponder these while I am setting up my new lab + home.

Fred.

I went back to the rubber bucket, this may be better, to me it is more musical.

The voice of the EWS Alive

This is only acoustically enhanced, no reverb or post EQ.

EQ could help tame the hot spots. I held the notes toward the end so Fred could evaluate.

Christopher

Hello Christopher,

EWS Alive is, IMO, a lovely sound... It  has some really 'acoustic' qualities ..

However - it does not have the (or IMO, any distinct) vocal qualities of your first sample. If this sample had been presented instead of the original, as a Cello-like enhancement, I would have congratulated you on its realism, but would not have been nearly as excited as I was on hearing the original sample..

Because, IMO, the original sample was not a Cello, it was a voice - close enough to the human voice to convey the emotional atributes that only human voice can carry, but different enough that it was not just a synthesised voice like the TM.. or at least that is my impression for the brief sample presented.

Electronic formant filters (particularly accurate digital ones like the TM) irritate me after a short time - I think the reason is something like this - they are accurate enough to fool me into hearing a person, but the downside of this is that I expect more from a human vocalist than these electronic 'voices' can ever deliver! - So, even in the hands of the best musician, they can never (or at least not with present control technology - perhaps in future when direct neurological or biometric connection - so that the performer can control the nuances of the sound..) be as good as even a mediocre singer.. I like good voices singing well - Given the choice of either hearing PP singing or hearing him play theremin through the TM, I would listen to him singing any day! -

So I seek a middle ground - formants which can convey vocal charactaristics, but fall just short of fooling me into expecting vocal proficiency.

Your first sample sounded like it may do this - it was too brief and (no insult intended) unmusical for me to be sure of how it would sound when played more proficiently and for a longer "piece" .. <edit>(in fact, after listening to it again, i wonder - it might even be too human for my ears! - I think perhaps, if those formants remain over a wider pitch range, it might just irritate me the way the TM does! LOL ;-)

 I think I "jumped the gun" a bit on your original posting / sample in the "Effects" topic http://www.thereminworld.com/Forums/T/24805/effects?Page=2 .. Sorry! you were just presenting an idea for achieving the requested cello sound, and not detailing the nature of the sound driving your resonating rubber bucket.. I know what its like to have loads of samples from old (often abandoned) experiments and not being able to remember exactly how one got that sound, or being unable to replicate it even if you think you remember how you got it originally...

If you got the vocal sound from an unprocessed theremin driving a rubber bucket or any other purely acoustic resonant setup without any pre or post processing, then it would be a real shame if you werent able to recreate this.

Personally, I really like acoustically imparted 'processing' - To some degree one always gets an acoustic influence from ones loudspeaker charactaristics and the environment (one of the reasons why I think vintage theremins need to be recorded with their original amp and loudspeaker, and that one cannot get their sound unless one was to take these into the equasion) - but transducers connected to resonators impart a whole load of complexity to the sound, and to my ears, even if one gets peaks or even buzzes, the sounds are usualy far more interesting..

And with theremins, adding a transducer which modulates / vibrates the pitch antenna IMO adds a whole extra sonic dimension -or  this can be a vibrating  plank of wood with grounded foil over it, positioned behind the antenna, or one can attach a transducer to the base of the antenna causing it to vibrate (this gives more modulation as the hand approaches the antenna - I prefer more modulation at the bass end) or just attach the transducer to the case, and have a antenna wire which is caused to vibrate with reference to some ground plane - this was the mistake which caused me to discover the effect in my H1 theremins which were built into a loudspeaker cabinet, with the antenna lead close to the port so it was moving about like crazy!

Fred.

* I should just add that by "modulation" I am not talking about vibrato or the like - one needs to drive the transducer at the natural difference frequency of the theremin - no register switching tricks! ... and the frequency shift at the VFO resulting from the vibration must be quite small - one only wants enough to distort the resulting waveform, and particularly produce 'irregular'  2nd harmonic distortion - other harmonics will be generated from resonances of the enclosure or whatever... When applying audio directly to modulate the VFO (without an acoustic resonator) increased 2nd harmonics are most noticable, but from an acoustic resonator they can be all over the place! ... Best to have a level control on the drive to 'tune' the sound... Taken after the VCA (normal theremin output) one gets more modulation as volume increases.