This will be hard, although Carmen's suggestions could be helpful. I
use Csound for stuff like this; it has everything you need to do it,
but it has a pretty steep learning curve. Actually I think when
people say that they mean a shallow learning curve, i.e., a slow one.
It's a programming language with every audio trick imaginable, and so
a lot to weed through to get what you want. It is less a program than
a programming language.
A spectral gate will analyze sections of the sound for frequency, and
remove frequencies with high levels or low levels or whatever levels
you want. Speech and Fourier analysis work strangely, though... A
Fourier analysis that analyzes a large block of sound will check
levels for many frequencies, but blurs changes in them. A Fourier
analysis on small blocks of sound tells you levels for fewer
frequencies, i.e. it has less frequency resolution, but it detects
changes in pitch and spectrum, like speech. Time resolution and
frequency resolution are basically indirectly proportionate.
Csound runs by default from the command line, but you have to write
files for it to run. This can be done with any text editor.
If you or anyone does check out Csound for this, the commands to look
at are "pvsanal" and "pvstencil". All the "pvs-" commands work with
Fourier analysis.
-Chuckk
On 3/9/07, terrence@xxxxxxxxxxxxxxx <terrence@xxxxxxxxxxxxxxx> wrote:
Hmmmm...well it's actually an old tape of all spoken stuff;
conversations I had while visiting family in Europe. I want to hang on to
the other people's voices etc., but mine has changed a bit since then and
the old sound bugs me.
Terrence
--
http://www.badmuthahubbard.com
