Re: [ecasound] ecasound versus ecatools and more

New Message Reply Date view Thread view Subject view Author view Other groups

Subject: Re: [ecasound] ecasound versus ecatools and more
From: Kai Vehmanen (
Date: Sun Feb 25 2001 - 16:52:02 EET

On Fri, 23 Feb 2001, Conrad-von-Soest-Gymnasium wrote:

> o work with ecasound directly. Sometimjes I used the normalizer for I
> fou= nd no better way to figure out a max-amplitude of a file and in
> the same = way amplify it. But I think this should now also work with
> ecasound. I'm =

Normalizing a file is not possible in ecasound, because it's inherently a
two-phase procedure. The first phase is to analyze the whole input file,
and then, as a separate procedure, amplify the input data (based on the
1st-stage results). While it's possible to specify multiple configurations
in ecasound (=multiple chainsetups), there are no mechanisms for
transfering information between then, nor ways to specify a chainsetup

If you don't want to use ecanormalize, you can of course do the
normalizing "manually" with ecasound. First you issue "ecasound -i
file.wav -o null -ev". Among the resulting statistics is ... "(audiofx)
Max gain without clipping, all: 1.2345.". You can then use this
information to do the normalizing, ie. "ecasound -i file.wav -o
file-normalized.wav -ea:123.45".

> e returning of values by the different commands and passing this
> results = to other commands and operators should make it possible. In

I think the best way to do this is to write a new application on top of
ecasound. This way ecasound doesn't get too complicated (= it is still
usable for simple tasks), while you can create as complicated interfaces
as you need.

> Then there is still my old worry about synchronizing tracks. The
> last t=
> t year. Maybe one could connect a noise-gates output to the starting
> comm= and. Does the noise-gate have the capability to output values as
> well as = some other operators. If so one could probably create a
> condition between the thresh-hold-level and the recording of
> audio-data to a file or at le ast setting a mark at the point where
> the gate opens. another

Hasn't been tested, but this should be possible with the new gate
system (1.9dev1 and newer). A simple example:

ecasound -i file.wav -o file2.wav -ge:10,0,0

... here, ecasound will start writing to file2.wav only after the
volume-level of file.wav has reached 10%. The second parameter specifies
the threshold for closing the gate, so we use 0% to ensure that gate is
never closed. The last param specifies the volume analyzing mode (0=peak,

> Another question that thrilled me for a long time is: could there be
> a = way to implement something like ecawave in console-mode? That is
> just a q= uestion, but I would be glad if you could think about this.
> Maybe you cou= ld just resume the main-functionality of ecawave, so

Actually most of ecawave functionality is in libecasound, so porting
ecawave to new UI-toolkits shouldn't be too difficult. But of course,
writing a wave-widget for console-mode won't be an easy task, but not
impossible either. Someone would just need to do it.

> A last small problem we had in handling .ewf-files with 1.8.5d15.
> This = was installed on a suse 7.0 linux system working on a laptop
> with a prope= rly installed alsa 0.5.8b I think. I tried the following
> first: ecasound -i test1.wav -o my.ewf That didn't work.

Upgrade to ecasound 1.9dev1 or newer. 1.8.5d15 had a serious bug in

 . ... [ audio software for linux ] /\ . 
 . ... [ aivastus net radio ] /\ .

-- To unsubscribe send message 'unsubscribe' in the body of the message to <>.

New Message Reply Date view Thread view Subject view Author view Other groups

This archive was generated by hypermail 2b28 : Sun Feb 25 2001 - 16:15:56 EET