In message , Bill in Co
writes:
[]
I did some more experiments. The pitch shift is around 40 cents, or a bit
over 2 %. I can definitely tell if a song is off pitch by 2 %. (In fact,
on the radio, if it's off that much, it's annoying to me (and I can tell
that from my own memory of the original song, even from years ago). It

Hmm, a third of a semitone, nice.

happens on all of the windows sound events, since they are all 22.05 kHz
wave files. And yes, it has always been this way on this one laptop.

So just for kicks, I converted one of them to 44.1 kHz, and sure enough, it
sounds fine; reconvert it back, and got the same problem.

And again, this issue happens on all my media players (with the exception of
that one simple one called Billy Player, which apparently must process the
22K audio signal differently - and correctly). (And it is nowhere near a 9
% pitch error, so that other idea isn't explaining it, either). Maybe it
has something to do with the buffering as Paul suggested, but I don't know
(some of these sound events are only one second)


So the fault only occurs with one set of hardware (that one laptop), and
at only one sample rate (have you tried 11.025 kHz? Or something that
_isn't_ a binary multiple/factor thereof, such as 24 kHz?), unless
played with one particular player.

I'm out of ideas (-:!

I think seeing if the _duration_ is affected might be useful to know,
although I'm not sure why. (You'd need a long enough file to be able to
time it, though, which for a 2% error, probably means a couple of
minutes or so at least. With a clearly-discernible start and end so you
know when to click the stopwatch or whatever.)
--
J. P. Gilliver. UMRA: 1960/1985 MB++G()AL-IS-Ch++(p)Ar@T+H+Sh0!:`)DNAf

Everyone learns from science. It all depends how you use the knowledge. - "Gil
Grissom" (CSI).