kanewolf :
Voice data is between about 200 to 4000 hz (telephone uses 300 to 3000). You need to minimize frequencies outside this range.
No no no no. While it's true that most of the voice frequencies fall in that range, it's the higher 4k-14k frequencies which help you distinguish one consonant from another. If you cut off those higher frequencies, you end up with a muddy garbled mess which is harder to understand. The only reason telephones leave out those higher frequencies is to reduce bandwidth. 200-4000 Hz is 4.3 octaves, so is pretty efficient in terms of bandwidth per octave. Encoding 4000-8000 Hz takes more bandwidth than encoding all of 200-4000 Hz, but only gives you 1 additional octave.
Here's a video showing the difference between regular voice calls and HD voice (50Hz - 7kHz). Those higher frequencies really help (they're a natural distance filter because these higher frequencies attenuate much more quickly in the air, so background noise contains very little of these higher frequencies). This is the same reason why a female voice is almost always a better choice than a male voice for your GPS.
https
/www.youtube.com/watch?v=LXEIi2JwgR4
It sounds like you don't have a 5.1 or 7.1 speaker system. What sort of A/V equipment are you using to produce the sound signal? If it's piping 5.1 or 7.1 sound through soundbar speakers, you may be including a lot of background sound which is supposed to be played only through surround speakers. If you're using an AV receiver, try switching it to stereo mode.