>from Lloyd... >It appears that you are assuming that the strength of the partial in >the phonated sound is the only possible source of strength in the >resonated tone.
Dear Lloyd, Yes, but the fundamental, I suggest is the repetition rate of closure and the all important harmonics are a function, as I imagine it, of the closing and opening _rates_ as well as the duty cycle. Put another way, I regard the speed of the transition from open to closed, and closed to open, to be crucial to the production of the harmonics and their amplitude. The duration for which the folds are closed, relative to the duration for which they are open, is another important and adjustable factor in the production of harmonics but NOT, of the fundamental. For the fundamental, any duty cycle, or closure rate will be ok, as long as the period remains the same.
> There is the physical phenomenon of an originating >tone being amplified acoustically by a resonating chamber.
Lloyd I'm sorry, you may have missed my submission some time ago that this is an unfortunate misconstruction of certain other physical phenomenon. Namely, that although resonant devices can be used to greatly enlarge signals, without careful buffering, the act of removing energy from them, will destroy the very characteristic we value so highly.
> Of >course, for this to happen the energy must come from somewhere and >the difficulty is from where? The resonating characteristic of a >closed tube is able, if conditions are correct, to borrow (as it >were) energy from dampened partials or partials that are not >resonated as strongly as their original strength upon entering the >resonating space.
We have to be careful here that we're not confusing purification by resonating with energy translation by intermodulation. A resonant device will normally dissipate as heat, in an external load, any signal that does _not_ fall within its tuned bandwidth. I consider that you correctly rejected intermodulation as a source of vocal tone some time ago, although it's difficult to be absolute on this principle. Still, there would be clusters of sum and difference signals surrounding the harmonics.
> Perhaps this energy is transmuted to provide the >increase in acoustic amplification in addition to that more naturally >supplied by the emphasis of particular partials.
To me, transmuted is another word for intermodulation, and if we were to accept that it occurs, _which I doubt_, the place I would be looking would be around that 6 :1 incline, but I don't see that as the grossly non-linear device that would be required to produce the mixing. Maybe! And if we took all the energy the early harmonics had to offer in the simple series, it would still be less than that of the fundamental, and that doesn't allow for conversion losses.
I can't accept the use of the word amplification here, because it implies active buffering of resonant cavities, and no such buffering exists beyond the larynx. This is critical to my proposition that resonance is used to purify and impedance match, and that the apparent enlargement, is actually an efficiency gain in combination with the moderation of temporal effects that enable a larger tone to be produced.
>There is also very strong evidence to indicate that the male voice is >incapable of producing a fundamental throughout its total range until >it reaches the frequency of about G4 and above.
I can accept this, because of the vocal tract dimensions and their use as part of the low frequency purification process, but I still suggest that including consideration of the effects of the 1/4 wavelength of the tract diminishes the problem. Apart from that processing difficulty, I cannot imagine why there should be any problem with a closure repetition rate at the lower frequencies, although the desirable open fold duration, may be difficult to maintain as we go still lower.
> What we hear is the >spectrum of partials that would normally be found with a given >fundamental and our ears (or minds) provide the phantom fundamental. >If this is so, then the calculations of relative strengths of the >partials would change appreciably.
Sounds ok to me :) I have no problem with the possibilities of aural integration of groups of high rate pulses at a lower cyclical rate. It offers the possibility of still greater harmonic energy if that's what we want. The calculations change drastically anyway as soon as we move away from the idealised 50% duty cycle which was only the starting point.
I cannot express how highly I value your input Lloyd and I'm sure that your vast knowledge and enthusiasm will be greatly missed following your retirement next fall. So I wish you a happy retirement, if that's what you intend. I am most appreciative.
Regards Reg.
|
| |