From: Kenn Humborg (kenn at domain linux.ie)
Date: Thu 20 Jul 2000 - 22:41:07 IST
On Mon, Jul 17, 2000 at 12:04:59PM +0200, McDaid, Aaron wrote:
> > > > what would be the best way to split this?
> > Ugh! Would split not give you a fscked MP3? It might play okay but the
> > format will be screwy.
>
> What about using 'mpg123 -s' to output the file as raw
> sound data (see the manpage) to a file, then use split etc.
> and then re-encode it as mp3.
>
> This would give a "valid" mp3 file and the raw data won't mind
> being split (although you may want to split every 2 or 4 bytes
> as opposed to any byte - see man page I suppose)
>
> Anyone know an easy way to listen to the "raw" file. You
> could maybe use head(1) and tail(2) as easier ways to find
> the parts you want to keep/lose
[This may already have been answered, but I've quite a backlog
of email to catch up on.]
Use dd. The default blocksize of 512 bytes should give fine
enough resolution. Then use the skip= and count= options to
specify the start and length of the chunk you want to extract.
Send this to /dev/audio to preview it:
dd if=raw-sound.au of=/dev/audio skip=1000 count=2000
Once you're happy either replace /dev/audio with a file and MP3
encode it, or remove the of= option and pipe it to your MP3
encoder.
Later,
Kenn
This archive was generated by hypermail 2.1.6 : Thu 06 Feb 2003 - 13:06:56 GMT