Hi,
I have made a patch which adds support for samplerate conversion to
XMMS's OSS output plugin. Samplerate conversion is done using Erik de
Castro Lopo's libsamplerate aka Secret Rabbit Code.
Why? Because there are soundcards which support only single samplerate
in hardware (usually 48 kHz). To get best possible sound quality out of
these, you'll need high quality samplerate conversion when playing 44.1
kHz files/streams.
All comments on this are welcome.
Patch is available from:
http://www.sonarnerd.net/linux/xmms-rabbit.patch
--
Jussi Laako <jussi.laako(a)pp.inet.fi>
Hello. I am a not-very-technical person trying to get some opinions to
pass along to our group.
We are trying to get speech recognition working on Linux and want to
test everything using the same USB sound pod and the best microphone,
provided that the equipment is widely available to the general public.
Right now the microphone of choice seems to be the Sennheiser MD 431 II.
For USB Sound Pod we understand that many of them do not run properly
with Linux, and of those that do, some do not transmit enough subtlety
to capture small variations in speech.
Any and all suggestions or comments would be appreciated.
Susan
Susan R. Cragin, Director
Open-Source Speech Recognition Initiative
susancragin(a)earthlink.net
http://www.ossri.orghttp://harvee.org/mailman/listinfo/ossri
susancragin(a)earthlink.net
liblo is an implementation of the Open Sound Control[1] protocol for POSIX
systems. It is written in ANSI C and released under the GNU General Public
Licence. It is designed to make developing OSC applictions as easy as
possible.
http://plugin.org.uk/liblo/
This release adds Mac OSX compatibility fixes from Taybin Rutkin, a
memory leak fix from Jesse Chappell and methods and examples to allow
server polling from exisitng threads from Sean Bolton. Some legacy
compatobility code has been removed, but this should not affect anyone.
Documentation has been updated to reflect the changes.
[1] http://www.cnmat.berkeley.edu/OpenSoundControl/ [website is currently
down]
Going slightly off topic, an idea I'd like to try is a hexagonal drum pattern
sequencer. I got the idea from this:
http://www.soundonsound.com/sos/apr04/images/autechremax2.l.jpg
I have no idea how it actually works, but from that patch I got thinking that
you could get multiple loops from the same pattern by moving time in different
directions across a (non rectangular) tesselated grid.
If you see what I mean!??
I'm going to hack together a little script to do this, when/if I have time,
but feel free to take this idea further...
On Tue, 19 Oct 2004 09:28:40 +0200, Aaron wrote
> So your saying the idea of rectangles might be patented??
>
> So I might be forced to use pentgons or octagons? Maybe the color
> will also be a problem....
>
> Aaron
> On Tue, 2004-10-19 at 09:09, Juhana Sadeharju wrote:
> > >From: Aaron <aamehl(a)actcom.net.il>
> > >
> > >1. Basically a front end to lilypond which will work
> > >more like a audio program.
> >
> > I remember something like that has been discussed somewhere
> > earlier.
> >
> > Well, if the rectangle on the track editor is made with audio editor,
> > it is displayed as waveform or as spectogram. If the rectangle on
> > the track is made with midi editor, it is displayed as matrix.
> > If the rectangle is made with X editor, it is displayed as X.
> >
> > When one zooms in to these displays, they might become editable.
> > (Waveform display usually does.)
> >
> > Maybe this should be mentioned in this context (before anyone
> > patents all these ideas): a rectangle may have different
> > reprentations and thus different display styles. E.g., if the
> > original data is audio, then the another representation could
> > be sequence data (midi?) generated with, e.g., wave-to-midi.
> > When the audio is edited, the other representation is changed
> > automatically (possibly using lazy-evaluation). If the original
> > is the sequence, then audio is just renderation of the sequence
> > (sort of freeze feature).
> >
> > I wanted to write about this multiple-representations because
> > in graphics, Silicon Graphics (Wavefront-Alias, Maya) has a patent
> > on using multiple reprentations. User-friendly for user-editing,
> > faster for rendering. I invented exactly that basic idea in high
> > school 10 years before they!
> >
> > So, if there are further ideas, lets hear them all now before
> > the ideas are patented.
> >
> > Juhana
>From: Aaron <aamehl(a)actcom.net.il>
>
>1. Basically a front end to lilypond which will work
>more like a audio program.
I remember something like that has been discussed somewhere
earlier.
Well, if the rectangle on the track editor is made with audio editor,
it is displayed as waveform or as spectogram. If the rectangle on
the track is made with midi editor, it is displayed as matrix.
If the rectangle is made with X editor, it is displayed as X.
When one zooms in to these displays, they might become editable.
(Waveform display usually does.)
Maybe this should be mentioned in this context (before anyone
patents all these ideas): a rectangle may have different
reprentations and thus different display styles. E.g., if the
original data is audio, then the another representation could
be sequence data (midi?) generated with, e.g., wave-to-midi.
When the audio is edited, the other representation is changed
automatically (possibly using lazy-evaluation). If the original
is the sequence, then audio is just renderation of the sequence
(sort of freeze feature).
I wanted to write about this multiple-representations because
in graphics, Silicon Graphics (Wavefront-Alias, Maya) has a patent
on using multiple reprentations. User-friendly for user-editing,
faster for rendering. I invented exactly that basic idea in high
school 10 years before they!
So, if there are further ideas, lets hear them all now before
the ideas are patented.
Juhana
--
http://music.columbia.edu/mailman/listinfo/linux-graphics-dev
for developers of open source graphics software
I'm currently doing this in TCL/TK. I started using the snack audio
library which (anong other things) makes spectrograms. I give it
consecutive chunks of audio and have it redraw to the canvas every n
milliseconds. I think it's having trouble keepingup, and it only knows how
to do 2d plots.
I'm considering doing the processing beforehand (ie, on load) and storing
the entire time-domain fft into a matrix, then using a graphics library
(plot3d) to graph it in real time... unless I can figure outhow to
multithread the program to allow one function to read ahead while the
other graphs it.
Anyway, I've never programed tcl/tk before, but I'm trying to get this
done before january.
-tewner
> On Sun, 3 Oct 2004, Dave Phillips wrote:
>
> > HiAndres:
> >
> > At last, someone's going to try it ! :)
> >
> > I'm hoping that you're thinking of a realtime display, in which the
> > peaks roll off to create a true waterfall effect.
> >
> > Stanko Juzbasic has tried to port Alan Peever's Spectrogram from SGI
> > machines to Linux, but I've not been able to get his sources built on my
> > systems.
> >
> > Please keep me informed about your progress, I've wanted such a
> > program for many years.
> >
> > Best regards,
> >
> > dp
> >
> >
> >
> > Andres Cabrera wrote:
> >
> >> Hi all,
> > > I am planning to develop a 3d fft display (sometimes called cascade
> > > display), since I haven't found an application that does that in linux.
> > > Does anyone know if such an application exists or is in the works?
> > >
> > > Cheers,
> > >Andr?s
> > >
> > >
> >
> >
> >
>
Hi All,
I have a small utility (jackswitch - a clickless <hopefully> audio switch for
the jack enviroment) and was wondering if there is a site for hosting this
kind of small simple utility program.
Sure I could create a web site for it but it seems rather silly to write a
website that is probably going to be larger then the tarball it is created to
host!
Steve, would plugin.org.uk take it?
