Hello all,
I spent a lot a time reading whatever docs I could find for Pipewire,
and discuss things with some users, only to get frustrated more and
more.
Below is a description of the configuration I'd want. If anyone knows
how to do this (it shouldn't be that difficult) that person will
receive my eternal admiration and gratitude.
1. Jack2 and some clients are started manually after I login,
and will be running all the time.
2. Currently the ALSA Jack plugin is used to route audio from
web browsers etc. to Jack. PW may take over this role but
that is not a strict requirement.
3. PW will be started manually when required, and I don't expect
that will happen very often. It may remain running when no longer
needed but shouldn't interfere. It will be used to connect apps
to Jack as in (2), or those that even don't support ALSA, or
maybe to route audio from Jack to Bluetooth etc.
4. All Jack ports created by PW should be permanent and exist
as soon as PW is started, so they can be manually connected
and remain connected even when not in active use.
5. PW should never ever access the sound card used by Jack,
not even if accidentally started when Jack is not running.
It must not force Jack to use dbus in order to get access
to that card. It may manage other sound cards, but preferably
only those explicitly listed.
6. PW must never ever interfere with Jack in any way - making
connections, trying to change the period size, etc. Its only
role is to be a well-behaved Jack client.
7. I do not expect anything 'automatic' to happen when things
are plugged in or out.
8. The PW configuration should be done in such a way such that
it can't be modified by drop-in files from the system package
manager. All configuration should be manual and explicit, and
easy to verify without having to scan a myriad of files and/or
directories and trying to understand how they interact. This
is just basic security.
Ciao,
--
FA
Ratatouille is a Neural Model loader and mixer for Linux/Windows.
This release does several small fixes.
Ratatouille allow to load up to two neural model files and mix there
output. Those models could be [*.nam files](https://tonehunt.org/all) or
[*.json or .aidax files](https://cloud.aida-x.cc/all). So you could
blend from clean to crunch for example, or, go wild and mix different
amp models, or mix a amp with a pedal simulation.
Ratatouille using parallel processing to process the second neural model
and the second IR-File to reduce the dsp load.
The "Delay" control could add a small delay to the second model to
overcome phasing issues, or to add some color/reverb to the sound.
To round up the sound it allow to load up to two Impulse Response files
and mix there output as well. You could try the wildest combinations,
or, be conservative and load just your single preferred IR-File.
Each neural model may have a different expected Sample Rate, Ratatouille
will resample the buffer to match that.
Impulse Response Files will be resampled on the fly to match the session
Sample Rate.
Project Page (source code):
https://github.com/brummer10/Ratatouille.lv2
Release Page (binaries):
https://github.com/brummer10/Ratatouille.lv2/releases/edit/v0.8
enjoy
hermann
Ratatouille is a Neural Model loader and mixer for Linux/Windows.
This release implement easier File switching. It's now possible to
switch the selected files via mouse wheel, via mouse button click and
via keyboard up/down keys. Right mouse button click will pop up the
file-list and allow to select a file directly.
Also, it implement lighter CPU usage for convolution (IR-File), on non
power-of-two buffer sizes, by using the multi threaded FFTConvolver engine.
Beside that the GUI was a bit overworked.
Ratatouille allow to load up to two neural model files and mix there
output. Those models could be /.nam files/ <https://tonehunt.org/all>/or
<https://cloud.aida-x.cc/all>/.json or .aidax files
<https://cloud.aida-x.cc/all>. So you could blend from clean to crunch
for example, or, go wild and mix different amp models, or mix a amp with
a pedal simulation.
The "Delay" control could add a small delay to the second model to
overcome phasing issues, or to add some color/reverb to the sound.
To round up the sound it allow to load up to two Impulse Response files
and mix there output as well. You could try the wildest combinations,
or, be conservative and load just your single preferred IR-File.
Each neural model may have a different expected Sample Rate, Ratatouille
will resample the buffer to match that.
Impulse Response Files will be resampled on the fly to match the session
Sample Rate.
Project Page (Source code):
https://github.com/brummer10/Ratatouille.lv2
<https://github.com/brummer10/Ratatouille.lv2>
Release Page (Binaries):
https://github.com/brummer10/Ratatouill ... s/tag/v0.6
<https://github.com/brummer10/Ratatouille.lv2/releases/tag/v0.6>
Ratatouille is a Neural Model loader and mixer for Linux/Windows.
This release fix handling frame buffer-sizes of arbitrary size, aka. not
power of two, for the impulse response engine.
Ratatouille allow to load up to two neural model files and mix there
output. Those models could be *.nam files <https://tonehunt.org/all> or
*.json or .aidax files <https://cloud.aida-x.cc/all>. So you could blend
from clean to crunch for example, or, go wild and mix different amp
models, or mix a amp with a pedal simulation.
The "Delay" control could add a small delay to the second model to
overcome phasing issues, or to add some color/reverb to the sound.
To round up the sound it allow to load up to two Impulse Response files
and mix there output as well. You could try the wildest combinations,
or, be conservative and load just your single preferred IR-File.
Each neural model may have a different expected Sample Rate, Ratatouille
will resample the buffer to match that.
Impulse Response Files will be resampled on the fly to match the session
Sample Rate.
Release Page:
https://github.com/brummer10/Ratatouille.lv2/releases/tag/v0.5
Project Page:
https://github.com/brummer10/Ratatouille.lv2
Hi
I revisited a old guitarix project (SpecMatch) and ported it to python3.
SpecMatch aims to compare two Sounds and generate a Impulse Response
File from the different.
Intentional it was developed to easier the process to recreate a
specific Sound within guitarix.
Now, this day's while we've NAM and AIDAX, there are better way's to do so.
Hence I've u-bounded it from guitarix and makes it a tool on it's own,
as there are still the need to add convolution to get the expected sound.
SpecMatch allow to load two Sound files, compare the Frequency
spectrum's of them, and generate a Impulse Response File from the
difference. So it enable you to get the missing bits.
A other use-case is to archive the "Full Impulse Response" of a
destination file by using the usual NAM trainer input file as source
file in SpecMatch.
I've posted some of my results, using it this way, as a show case, on
the linuxmusicians forum her:
https://linuxmusicians.com/viewtopic.php?p=168587#p168587
This is, after all the years, still work in progress, and, there
shouldn't ever been a release to be expected, as it is plain
development. Anyway, if this stuff is of some interest for you, here it is:
 https://github.com/brummer10/SpecMatch
regards
hermann
SpectMorph 1.0.0-beta2 has been released.
Compared to beta1, the only change is that the plugin user interface now
works properly on macOS 14. You can get the new version from
https://www.spectmorph.org/downloads
If you do not use macOS 14, there is no reason to update.
Changes in beta2:
-----------------
* Plugin user interface now works correctly on macOS 14 (#28).
* Update clang++ compiler version on macOS.
* Minimum supported macOS version is now macOS 11.
Feedback for any issues you might experience with the beta version is
appreciated.
--
Stefan Westerfeld, http://space.twc.de/~stefan
