5 Jul
2004
5 Jul
'04
11:09 a.m.
hello list,
is there a way to normalize with sox in one pass? from the manual it seems
that you must first find out the max value, and then pass 32768 (or
whatever)/max to the vol effect...
best,
lj
--
5 Jul
5 Jul
11:18 a.m.
On Mon, 5 Jul 2004 10:01:57 -0300
luis jure <ljure(a)multitel.com.uy> wrote:
hello list,
is there a way to normalize with sox in one pass? from the manual it seems
that you must first find out the max value, and then pass 32768 (or
whatever)/max to the vol effect...
Normalizing _must_ me done in two passes.. That's the nature of normalization. Or do
you mean to ask if there's a way to call sox only once and it does the two pass
normalization itself? If that's the case: i have no idea..
But you could always write yourself a shell script to do that..
flo
--
Palimm Palimm!
http://affenbande.org/~tapas/wiki
6 p.m.
el Mon, 5 Jul 2004 15:15:24 +0200
Florian Schmidt <mista.tapas(a)gmx.net> escribió:
Normalizing _must_ me done in two passes..
thanks everybody for your answers.
yes, i'm aware that normalization must be done in two passes... i didn't
word my question very well, what i meant was invoking sox just once. i think
i'll either write a script, or else try "normalize". i'm not bound to
sox,
one of the students asked me how to normalize with sox, and reading the
manual sox didn't seem to be very clever, using the "stat -v" effect only
works when doing a conversion and needs an output file, at least that's what
i understood. writing a temporary file just to find out the peak amplitude
seems rather silly, but i thought may be i was missing something. i might
try my limited bash scripting abilities to write a nice script using
sndfile-info, grep, sed and sox. or else install normalize. we are not
using the eca-family, so i'll leave that one for now.
best,
lj
--
6 Jul
6 Jul
5:38 a.m.
[about sox]
using the "stat -v" effect only works when doing a conversion and needs
an output file, at least that's what i understood. writing a temporary file
just to find out the peak amplitude seems rather silly
if this is indeed necesary, you can always use /dev/null as the output file.
it still seems rather silly, but at least you won't need the disk-space for
the temporary file, and it will be faster too, writing to the sink instead
of to the harddisk.
maarten
7:48 p.m.
On 05-Jul-2004 luis jure wrote:
} word my question very well, what i meant was invoking sox just once. i think
} i'll either write a script, or else try "normalize". i'm not bound to
sox,
http://www1.cs.columbia.edu/~cvaill/normalize/
Why would you not just do this in the first place?
----------------------------------
E-Mail: RickTaylor(a)Speakeasy.Net
Date: 06-Jul-2004
Time: 16:35:56
This message was sent by XFMail
----------------------------------
5 Jul
5 Jul
11:33 a.m.
If you need to normalize you should check out "normalize" -
http://www1.cs.columbia.edu/~cvaill/normalize/ If you want to master a
project you need to check out JAMin - http://jamin.sourceforge.net/ ;-)
Jan
On Mon, 2004-07-05 at 08:01, luis jure wrote:
hello list,
is there a way to normalize with sox in one pass? from the manual it seems
that you must first find out the max value, and then pass 32768 (or
whatever)/max to the vol effect...
best,
lj
11:43 a.m.
On Mon Jul 5 10:01:57 2004 luis jure wrote:
is there a way to normalize with sox in one pass? from
the manual it seems
that you must first find out the max value, and then pass 32768 (or
whatever)/max to the vol effect...
Are you bound to sox?
You could use ecanormalize. See ecatools(1).
--
Jan Weil
http://linuxaudioblog.jawebada.de
11:45 a.m.
On Mon, Jul 05, 2004 at 10:01:57AM -0300, luis jure wrote:
hello list,
is there a way to normalize with sox in one pass?
Maybe not with sox, but there's a program called (surprise!) normalize
which does what you want if the files are .wav or mp3.
It still works in two phases, but automatically.
--
Anahata
anahata(a)treewind.co.uk Tel: 01638 720444
http://www.treewind.co.uk Mob: 07976 263827
7492
days inactive
7493
days old
linux-audio-user@lists.linuxaudio.org
7 comments
7 participants
participants (7)
-
Anahata -
Florian Schmidt -
Jan Depner -
Jan Weil -
luis jure -
Maarten de Boer -
RickTaylor＠Speakeasy.Net