Put the phones on, up close on the mic now, do a nice low Bing Crosby swoop'.

As I understand it the phones block the air path to the ear and the path from the mic can be out with the path through your head. (In the lows mostly.
Ahhh, you're talking about live monitoring of your own singing. I missed that. OK, I can see that. There you *do* have two sources, one's own voice through bone conduction, and one's voices as heard through the mic path. And yeah, I can see where there would be potential phase cancellation issues there. However, If Rami is just listening to playback of a recording (or live from someone else singing), polarity inversion in and of itself should have no audible effect at all, even in cans.
Phase in fact is considered as a shift in time, so when you have a 180 degree phase shift you in fact have the opposite polarity.
True only if there is no DC offset in the signal. Granted, 99 times out of a hundred, there won't be offset (unless you're going through a stock laptop soundcard, in which case all bets are off

.) But, the real fact is that a 180° phase change means a change of phase angle around the signal's rest voltage, whereas a polarity inversion means a similar change of phase angle around 0DC. Just because a lack of DC offset makes them appear the same, does not make them the same - or at least they will be measurably the same only if there is no offset. But just throw even a millivolt of offset into the signal, and they will differ.
But polarity is in fact other thing because for example, lets say you have 2 sine waves and you change the phase 180 degrees on one of them, youll be ending with the opposite polarity of one of the sine waves with respect to the other but there will also be a delay on one of them because theres a time shift.
This is not necessarily true. What you are describing is a phase shift, or - as I personally prefer to call it - a phase rotation. This rotation happens in time, which does indeed, by definition, make that a time-dependent process.
But it is entirely possible to change phase independent of the time variable; i.e. to have instantaneous phase change without having to shift position in time. Granted, not by sliding a waveform around on a timeline. But by executing a simple formula V
2 = -V
1 + Vr, where V
2 is the new voltage value, V
1 is the original voltage value, and Vr is the signal rest voltage or offset.
G.