Quote:
|
Unless I'm using my "bass voice" I typically don't get down below 300 or 400 and I'm a baritone.
|
Actually, you do. It's getting off topic. But A440 is the equivalent of the 5th fret 1st string on a guitar (A4). 300 Hz is just about D4, the 3rd fret on the 2nd string. Your bari voice likely goes well below an octave below that to perhaps low G2 (about 100 Hz, 3rd fret 6th string) or the textbook classic baritone, low E2 (about 82 Hz, open 6th string)
It's my understanding that even POTS is no longer limited to 300-3k, the old limits of the copper wires imposed due to load managment. Cellular has a higher audio bandwidth.
It is accurate to say that it's not quite as obvious to tell one voice from another in spectral analysis when the bandwidth is limited. But it's certainly not impossible or even very difficult. The freq response of each individual voice (which frequencies are louder or softer than others) is very constant whether we're talking, singing, grunting, screaming. Even with a very poor signal to noise ratio (background noise, hiss, bad phone connection) it's a relatively simple matter to say "This sample voice spec analysis looks like that sample - but it doesn't look like that other sample".
Sgt Lumpy