Rock Tone ML is a simple tool that takes dry recorded guitar signals and performs rock amplifier modification to the input signal. This is done using a PyTorch implementation of ML algorithms and neural network models that are trained to mimic the audio modification behaviour of amp replication software.
Using this tool, you can apply pre-trained tones to your recorded guitar playing. At the moment only one effect behaviour is being implemented as a proof of concept. Future plans do include adding the ability to learn from sample audio and for Rock Tone ML to mimic that tone on newly recorded audio samples.
This Project is currently work-in-progress, with the hope of having a tool release version available soon.
TODO
TODO
TODO
TODO
TODO
The current plan ahead for this project is the following:
- Researching the applications of several deep learning models on audio signal data (What is useful and what is not)
- Implentation of a virtual guitar amplifier that provides pre-cloned tones that could be applied to dry guitar signals.
- Implementation of a custome cloning function.
- Custom cloning would occur if sufficient hardware is available and a custom model can be trained.
- User would provide dry signal samples and desired signal samples before cloning can start.
- Creation of a front end that can simplify usage of the tool itself.
# Include specific commands and code snippets here