SmartGuitarAmp is a guitar plugin built with JUCE that uses neural network models to simulate real hardware.
This plugin uses the WaveNet model to recreate the sound of real hardware. The current version models a small tube amplifier, with the possibility of adding more options in the future. There are two CLEAN channels, which is equivalent to the clean channel on an amplifier, and a second LEAD channel, which is the distortion channel. Gain and EQ controls have been added to modulate the modeled sound.
You can also create your own models and upload them to SmartGuitarAmp. Training is performed using PyTorch on pre-recorded .wav samples.
Source: linux.org.ru