Subject: Re: [ecasound] Trying to normalize two channels independently
From: Andy Loening (email@example.com)
Date: Sun Nov 05 2000 - 20:55:22 EET
Allright, after wasting two weeks building a new computer and trying to
get Redhat 7.0 to compile software (avoid, I'm back to 6.2), I return with
my original question.
Just to refresh everyone's memory, I have a live audience recording where
I'd like to normalize the left and right channels independently to utilize
the full 16 bits of dynamic range (the left channel needs ~133%, the right
So, I grabbed the cvs version of ecasound/etc, and threw it the following
commands (doesn't crash now, yeah!).
#to convert to two channels
ecasound -sr:48000 \
-a:1,2 -i 2nd_encore.wav \
-a:1 -f:s16_le,1,48000 -o channel_1.wav \
-a:2 -f:s16_le,1,48000 -o channel_2.wav -erc:2,1
#and then normalize each channel
#to combine back into one channel
ecasound -sr:48000 \
-a:1 -f:s16_le,1,48000 -i channel_1.wav \
-a:2 -f:s16_le,1,48000 -i channel_2.wav \
-a:1,2 -f:s16_le,2,48000 -o try_again.wav
It all seems okay up until the last command where I'm trying to convert
the two mono channels back into a single stereo file. When I use the
command as written above, I get both incoming channels on the left channel
of the output file.
Can anyone tell me how I could get the mono channel from channel_2.wav
(the original right channel) to show up on the right channel of the output
I also have a question about ecanormalize. Is ecanormalize going to
internally resample to 44100 when it's doing it's thing?
On Fri, 20 Oct 2000, Kai Vehmanen wrote:
> On Tue, 17 Oct 2000, Andy Loening wrote:
> > ecasound \
> > -a:1 -i channel_1.wav \
> > -a:2 -i channel_2.wav -erc:1,2 \
> > -a:1,2 -f:s16_le,2,48000 -o new.wav
> > When I try to run this command, however, I get a coredump. The output of
> > ecasound is as follows:
> Ok, this is a bit messy one. First of all, the above command is correct,
> but not necessarily what you want. Problem here is that if you haven't
> touched your ~/.ecasoundrc, the result of the above command is that your
> data is resample twice. This is because ecasound converts all input data
> to its internal format (default sample rate is 44100) before any
> processing. So you either need to add a -sr:48000 to the command
> line, or change the default in ~/.ecasoundrc.
> But it was very good that you reported this, as it helped me to find _two_
> bugs. The first one was that -sr options has apparently been broken since
> 1.8.3d15. The second bug was in ecasound's resampling routines (caused the
> core dump). Both are now fixed and in the CVS-tree.
-- To unsubscribe send message 'unsubscribe' in the body of the message to <firstname.lastname@example.org>.
This archive was generated by hypermail 2b28 : Sun Nov 05 2000 - 21:08:29 EET