<font color='#000000'>In the dim recessess of my memory I recall some "quasi-objective" listening tests one can do to weed out some typical speaker defects:
1)Artificially "forward" in the midrange. (a peak in the midrange response that seemingly gives a speaker an initially appealing "presence" or "sparkle" but is ultimately fatiguing. Or it can make the speakers sound "nasal", especially on female vocalists). Tune between FM stations for the hiss, or play white noise from a test CD. It should sound a lot like you hissing through your teeth (go ahead, hiss through your teeth!) If on the other hand it sounds like you have your hands cupped over your mouth while hissing (yeah, try it), then the midrange is crap. Oh, I think when using FM noise I used to turn down the bass tone control since there's a lot of low frequency noise there too, more like pink noise. You want to concentrate on the proportions of mids and highs, hence the use of white noise or its approximation with an FM tuner.
2)"Boomy" or artificially enhanced bass (typically around 100 hz). Listen to male speaking voices, especially deeper baritones and basses. If there is an artificially "chesty" or resonant/boomy quality, walk on by. Some people have also characterised the sound as if one were speaking into a large tube.
3) Excessive sibilance. If the "s" sounds in vocals (called sibilants) sound hissy, the high frequencies are too accentuated. Everything will sound overly bright. Again, the overall effect can be appealing on first hearing but will wear on you with extended listening.
I have also heard of a "stand up/sit down" test with pink noise to test off-axis vertical response. The less change off-axis the better, I guess. Horizontal off-axis quality is harder to hear, since on most speakers the biggest change is way up in the treble. And you have to test with just one speaker. Vertical off axis differences are often worse than horizontal when measured but, paradoxically, they're probably less important unless both your floor *and* ceiling are highly reflective or you spend a lot of time listening either laying on the floor (slacker!) or suspended from the ceiling (perv!).
Finally, the respected speaker designer Siegfried Linkwitz suggests putting the speakers in one room and listening from another room. The more it sounds like there are real people singing or talking, and real musicians playing their instruments in the other room and the less like loudspeakers, the better. I think this presupposes the use of well recorded acoustic performances and spoken word as source material.
Also, I would say that the best aural training is to attend and listen closely to as many *non-amplified* performances as possible. This will acquaint you with the actual sound of musical instruments and voices in a performance space. Tough to do if you're a heavy metal or hip-hop fan, I know...hmm, has System Of A Down ever done an "Unplugged" CD...? ;-{)}
(Old guy trying to sound hip. Humor him.)</font>