Re: [ecasound] Ecasound for long continuous recordings.

From: Kai Vehmanen <kvehmanen@email-addr-hidden>
Date: Tue Sep 11 2007 - 23:34:04 EEST


On Wed, 5 Sep 2007, Ricardo Antunes wrote:

> I'm looking into an open source command line sound recorder to use in an
> embedded linux system, to record sound files continuously for long periods.
> Ideally, the recorder would split the sound files into smaller,
> pre-determined size, files. Ideally no samples would be lost when
> switching to a new file, and file name would be some kind of timestamp
> (e.g. system clock time).
> I'm new to Ecasound, so not sure if it would be possible to use it for
> this application, but decided to check before starting to code something
> to do it myself.

at the moment ecasound cannot do this. People have asked for this many,
many times, so it would be great to have the feature sooner or later.

Technically, the solution could be something like "resample" and "reverse"
objects that provide a "added-value" wrapper for other audio objects. So
you could have something like:

sh> ecasound -i jack -o splice,foo_%v_%t.wav,200

This would:
   - record to file foo_%v_%t.wav from JACK
   - every 200secs, switch to a new file
   - in filename, %v is the serial number (1...N) while
     %t would be the cutoff (0 for first, 200 for next, and so on)

Any volunteers to implement this? :) The ewf file format implementation
is probably a good starting point (in addition to "resample" and

I'm currently working on a test tone generator for ecasound (aimed to help
reproducing complex issue/bugs without the need to create lots and lots
test input files) so I'm going to have my hands full for quite a while.

  links, my public keys, etc at
This email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
Ecasound-list mailing list
Received on Wed Sep 12 00:15:06 2007

This archive was generated by hypermail 2.1.8 : Wed Sep 12 2007 - 00:15:07 EEST