It's a nice and hacky solution because I was in a hurry and I know the
box that'll run the code.

So, I use ProcessBuilder to spin out a sequence of:

mpg123 infile.mp3 infile.wav
sox infile.wav tempo.wav tempo 1.2 30
lame -V 6 tempo.wav tempo.mp3

(on debian. I tried custom-building sox so that sox can read and write
mp3s, but no luck).

then I use the jid3lib library to copy the ID3 tags from infile.mp3 to
tempo.mp3.

Then I update the atom, and upload the whole shebang to a kind soul's
account who has graciously offered bandwidth.

I'll check if he wants his name spilled here on the board.

On Feb 1, 8:35 pm, Ed G <[email protected]> wrote:
> Hi Reinier-
>
> What filters do you use? I haven't looked at API-level filters, but
> for stand-alone apps, I haven't found many that will do this aside
> from Audacity. I use Audacity on WinXP to do this, and the new 3.x
> beta version has a batch-run feature. I set up a batch job that just
> compresses a mp3 and then saves it, so from the GUI I launch the
> batch, select a bunch of files, and then 10 minutes later they're all
> converted. Unfortunately I haven't figured out how to script it (or
> cracked open the code to see how it does it) to automate it.
>
> Regards,
> -ed
>
> On Jan 29, 6:52 pm, Reinier Zwitserloot <[email protected]> wrote:
>
> > There are some fairly simple audio filters that will speed up the
> > entire podcast without changing the pitch. (No chipmunks)
>
> > I'm willing to code up a bot that will watch the posse RSS feed,
> > download it, run the speedup filter, and then broadcast its own RSS,
> > no more than an hour delayed.
>
> > Assuming there are open source command line tools or a java library to
> > add chapter marks, the bot could get an email address or simple web
> > form where known contributors drop chapter lists, so that the bot can
> > automatically update the audio file it broadcasts. There'd then be a
> > second RSS feeds that would be delayed more, but always has chapter
> > markings.
>
> > I've mailed the posse about it. I can build it, but I don't have the
> > bandwidth to host this bot for more than about 75 listeners.
> > (Bandwidth bill where my server is at would become a bit too crazy to
> > pay out of my own pocket).
>
> > If anyone has the bandwidth to run it, let me know. It would probably
> > be a java servlet. I can set it up so that my home server does the
> > encode (saves your production server from having a chunk of its CPU
> > eaten away during the computationally intensive audio speedup). Reply
> > here or mail me.
>
> > On Jan 29, 3:56 pm, Jess Holle <[email protected]> wrote:
>
> > > Hmm... The only speed up I'd like is for the Posse to move on more
> > > quickly when they /know /they don't know what's going on -- rather than
> > > spending a long while saying as much and providing wild and crazy guesses.
>
> > > That said, I'm not complaining.  They're doing this of their own
> > > volition and are free to blabber on as they see fit.
>
> > > --
> > > Jess Holle
>
> > > Ed G wrote:
> > > > How about offering a speeded-up version for those who like the amount
> > > > of information but want to do it quicker? A 10% or 20% speedup would
> > > > not be very noticeable (not like chipmunks at all!) and would save
> > > > that much time. After a week or two of training your ears/brain I have
> > > > listened to podcasts at 30-40% speedup no problem, get almost double
> > > > the listening in!
>
> > > > On Jan 26, 9:54 pm, "[email protected]"
> > > > <[email protected]> wrote:
>
> > > >> Personally, I'd prefer a longer podcast but a lot of it has to do with
> > > >> the great signal to noise ratio this podcast has.
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "The 
Java Posse" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to 
[email protected]
For more options, visit this group at 
http://groups.google.com/group/javaposse?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to