NeuralRack is a Neural Model and Impulse Response File loader for Linux/Windows.
It’s a mono in stereo out plug.
Neuralrack allow to load up to two neural model files ( been it NAM or aidax models) and run them serial.
The input/output for each model could be controlled separate.
For tone sharping a 6 band EQ could be enabled.
Additional it allow to load up a separate Impulse Response file for each output channel (stereo).
Neuralrack provide a buffered Mode which introduce a one frame latency when enabled.
It could move one Neural Model, or the complete processing into a background thread. That will reduce the CPU load when needed.
The resulting latency will be reported to the host so that it could be compensated.
It does not seem to do anything that I noticed on LV2. It seems to maybe disengage SOME of the amp sim somehow? But not all I am using AIDX files . I did not try VST3
This is “normalize”. It works only when the model provide information about the expected output calibration factor. aidax models didn’t provide this info, and older nam models as well didn’t provide it. In that case it does indeed nothing.
For IR files in works in any case, it reduce there the influence on the loudness, so switching IR files becomes more smooth.
The first one, “app” is a stand alone application.
The second is aLV2 plug without optimisations applied, it’s for older distributions/CPU’s which didn’t support the x86-64-v3 optimisation level.
The last one, “v3” comes with all optimisations applied, it use lesser CPU then the second version, but required a modern distribution and CPU. To checkout beforehand if it will work on your system you could run /usr/lib64/ld-linux-x86-64.so.2 --help 2>/dev/null | grep 'x86-64-v3 (supported'
if that returns nothing, you could only use the second one.
I’m sorry. But github have skipped support for building against ubuntu 20.4, minimal version is now 22.04 which requires already glib 2.35
The standalone version still is build against 24.04 with all optimisations applied.
So your only option will be to build it from source.
There is a container image for 20.04.
I know it’s a bit of work but maybe consider in the future migrating to a build in containers.
I think you might also not need the “Fix GitHub’s mess” stages at that point either.
