(In progress)
This repository contains code for a deep learning approach to programming the Yahama DX7 synthesizer. This project is part of the Stanford Deep Learning Course CS 230.
The Yamaha DX7 is a classic FM synthesizer, known for being notoriously difficult to program. The goal of this project is to develop a system that can use infer from a raw audio input, the best DX7 parameter set to use to allow the recreation of that sound.
For generating data and testing our system, we are using the open-source
DX7 emulator dexed
.