Making music with Realtime Audio Styletransfer #WIP


I spent a few days at my parents’ house during the new year’s break. I had plenty of time, OP-1, headphones, and a great working environment there during the break. So I decided to make a track using the RAVE neural synthesis model trained on speech data with the real-time plugin!! 🎙

I fed drums and synth sound into the real-time RAVE plugin to transfer the timbre into speech this time. I found it interesting how the model interpreted the incoming audio as speech.

RAVE model
On RAVE VST plugin