I feel like this is a lot more approachable with the advent of cheap mass-produced RC helicopter parts (modulating the collective pitch with the audio waveform).
(Which has me wondering if you could do some fun directional things with the sound? Of course low bass is not exactly known for its directionality)
Sound with a wavelength of 6 inches is 2.25 kHz, a high pitched tone far above typical voices, around 100-250 hz. Pretty sure the ears can measure the delay between peaks in the waveform with wavelengths much longer than the width of a head.
That is true, in fact directionality is actually harder to achieve in higher tones (everyone remember those mosquito ring tones that went around a little over a decade ago?) When the wavelength gets very small your ears have less and less information to distinguish which ear is receiving the signal first, other than the brute force which ear actually heard it first calculation. Distinguishing waveforms becomes near impossible at high enough frequencies. It's interesting though that a low enough frequency might also be difficult to distinguish due to there being not enough information rather than too much. If the waveform is as long as a room how much difference is there really across the size of a head.
(Which has me wondering if you could do some fun directional things with the sound? Of course low bass is not exactly known for its directionality)