[hugo_zair]

It's essential to get a good understanding of this issue. Here's my take on it. Firstly, the concept is most clearly understood by considering plain tones (sine waves as illustrated in Drago answer above). Unfortunately speech and music don't compare too well with sine waves.

It's got to do mostly with loudspeaker and mic connections but also possibly with stages of gain in amplifiers. Taking the last one first, many individual gain stages within a pre-amp or other system channel will possibly flip the phase by 180 degrees. This should be taken care of by the designer who would have designed the whole circuit to present the same basic phase on the output and was received on the input. Unless you build you own part of a signal chain you shouldn't need to worry about that.

Next up in the "phase flip by 180 degrees" is loudspeakers and loudspeaker drive units (if building your own) If a pair of loudspeakers are wired opposite phase, there will be cancellation of the bass frequencies and poor stereo performance. To test this out place the two speakers facing each other and an inch or so apart, then reverse the phase of one with the cable connection. The difference in volume, especially in the bass output, is obvious .

Last is the question of mic phasing, which I guess is what the OP was referring to. The convention is that positive pressure (wave front) on the diaphram produces a positive going electrical signal on pin 2 of the XLR. But the issue of a phase error of 180 degrees is rarely relevant in this case except where there is a physical wiring problem. In use, it would be most relevant in the case of two microphones fixed very very close together as in a XY stereo pair. Here the effect would be diastrous and would have the same result as having the monitor speakers phase reversed. However it could be totally fixed by pressing the phase invert button on one mic.

In all sitiuations, and assuming the mics (and any cables in the circuit) match the convention,the most common battle with phase errors is with multiple mics responding to the same source from different distances; ie bleed/spill

The phase errors intoduced in this case vary with frequency and distance apart so, for example, the signal on one mic, when mixed with the bleed from another one short distance away wll produce all sorts of phase errors - called comb filtering. Some frequencies will tend to add and others subtract. But only a few frequencies may have an error of 180 degrees that can be properly fixed with the phase button.

What will happen is that , as Brandobury says above it will, SOUND DIFFERENT and it's a judgement you have to make as to which sounds best.

So..The phase reverse can only completey fix a problem in a mic or its wiring where there is a 180 degree phase error. But you shouldn't have any in your kit!

All other phase errors resulting from the use of multiple mics (the bleed or spill issue) at the same time, are random and vary with distance and frequency. The more mics you use the more skilled you need to be with critical mic placement and common advice is to use as few mics as possible.

There is a lot of good advice on the Shure website, here's a clip:

Audio Problems caused by Multiple Open Microphones.
High quality audio becomes progressively more difficult to achieve as the number of open microphones increases. All audio systems face the same problems whenever multiple open microphones are needed. These problems are:

* Build-up of background noise and reverberation
* Reduced gain before feedback
* COMB FILTERING (the relevant issue here)