Yes, "dodgy" is a good adjective for what you have done. As a result
of ignoring Nyquist's sampling theorem, the new code wraps (aliases)
frequency content originally at 34 to 44 kHz down to 34 to 24
kHz. What is needed is an anti-aliasing filter. I'll see what I can do
and mail you the results later on today.
Michael Cheng writes:
> Hi all,
> I'm trying to get some of these ideas out of my system before I
> sit down to do assignments for the next 2 weeks, so I implemented a couple
> of small things that I thought were neat.
>
> - new command switch "-a" resamples a stereo input file to mono
> - new command switch "-r" resmaples from 44.1khz to 32khz [this
> one is a bit dodgy, as the resampling produces some ringing sounds.
> Anyone who has had experience in resampling, have a look at what I did in
> encode.c]
>
> With these two switches active, you can encode at about a third of the
> speed of a stereo 44.1khz file (but you only produce a mono 32khz mp3)
> The sprdngf1/sprdngf2 functions for 32khz could probably be adapted from
> the ones I did for 44.1khz.
>
> later
> mike
>
>
>
--
James 'Jasha' Garnet Droppo III [EMAIL PROTECTED]
Interactive Systems Design Lab ISDL:(206)543-7298
Department of Electrical Engineering, University of Washington
http://isdl.ee.washington.edu/isdl/jdroppo/
