I have a wireless card in my PC connecting to a wireless router. No matter where I place the antenna for the card, I get about 30% signal strength.
When I place the antenna on my PC speakers, it jumps to 50%. I have fiddled around a bit with positions, but no matter what, my signal jumps up by 20% when placed on the speaker.
If I switch the power to the speaker off, it drops be 20%
Reason dictates that, by putting your antenna on a speaker (instead of just the desk), something to do with the speakers (which are? (make/model)) themselves is influencing the gain/boost you're getting. Going by your report, if the difference is that profound between the power being on/off, the gain doesn't come from getting a better
groundplane from any nearby metal component alone; evidently the presence of a electro(magnetic) field when the speakers are powered is the primary cause.
Now, seeing as the antenna on that is a fixed part and you don't get a free power boost for/from nothing, it
HAS to come from somewhere! Also, the antenna you're using is likely a little 3dBi
dipole on a base on the end of maybe around a metre of co-ax (similar to
this example) so there's no monster power boost from that source.
So, for me, it really seems to come back to the presence of a field that's causing the boost. Hence, going by a
certain famous maxim of deduction, that's the conclusion at which I must arrive - now, if only I could explain
why! (it's been way too many years since I last looked at antenna theory, what I've come up with here has more to do with thinking things through than with what little I remember of anything I may have learnt way back when!) But, what little I can dredge up tells me that dipoles benefit from a better "ground" - I think I vaguely remember that from my CB (yes, I'm an oldie!) radio days when all I used was a dipole outside my window and I found it's effectiveness to differ with height.
So, HTH; if not, go and bother a radio communications engineer or something!