On Sun, May 18, 2008 at 5:39 PM, Al Thompson wrote:
I have had success with making multiple hours long sound files with
the soxrec sox frontend and with csound. I recall a pd bug with
especially long sound files, caused by using floating point as the
file seek index, but this may be fixed by now. Since these programs
only operate on the current buffer when creating sound files, they
don't have explosive ram/cpu usage problems, the way that a program
that updates a waveform display or stores the sound in ram as you are
creating it can. From my experience with very large recordings, your
most convenient option is actually a series of shorter sound files,
with a few seconds of overlap on each adjacent pair. This speeds
load/save/effects times, and also prevents exposing those corner case
huge file bugs that so many otherwise reliable apps are prone to have.
Making the overlapped files may be hard without jack... perhaps you
would need to use dmix for the input device, or whatever the
equivalent would be? I am not sure about that part. I will see if I
can find the shell script I wrote to do this. Also using sox with the
trim command can make a series of shorter files from a mega file for
editors to work on.
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user@lists.linuxaudio.org
http://lists.linuxaudio.org/mailman/listinfo/linux-audio-user
LINUX® is a registered trademark of Linus Torvalds in the USA and other countries.
Linuxaudio.org logo copyright Thorsten Wilms © 2006.
Hosting provided by the Virginia Tech Department of Music and DISIS.