Max Graf — Machine Learning Engineer on the CORPUS project and PhD in AI & Music from the Centre for Digital Music, Queen Mary University of London. XR musical instruments, hand tracking, interactive machine learning, and audio ML.
SA3 Variations generates Splice-style sample variations entirely on your machine. Drop in a one-shot or a loop and it produces a series of musical and timbral variations. Listen through the outputs, then drag the ones you like straight into your DAW. A batch takes around 10 seconds, runs fully on-device, and nothing is ever uploaded.
Stable Audio Remix is a small web app built around Stable Audio 3’s audio-to-audio mode. You drop in an existing track, describe the style you’d like instead, and it returns a version that holds onto the original’s timing and shape while moving it toward the new direction. There’s an A/B player built in so you can flip between the source and the remix and compare them easily.
WavNav is an audio sample explorer built to make large sample libraries easier to navigate. Instead of digging through folders and filenames, it maps sounds into a 2D visual space where similar samples sit near each other, so browsing becomes much more like exploring a map than managing files. If you want the full product overview, check out the WavNav product page first. It is much more in depth…
Demucs WebGPU is a browser-focused extension of sevagh’s demucs.onnx, which brings the Demucs v4 hybrid transformer into an ONNX-based inference pipeline for music source separation.
This project is a music analysis pipeline I developed as part of my work on the CORPUS project. Its purpose is to take incoming music audio and turn it into rich, structured metadata that can support browsing, filtering, moderation, search, and other downstream creative or editorial workflows.
Reef DJ is a proof-of-concept developed within CORPUS to explore a lightweight, interactive approach to AI music generation.
Netz is an XR musical instrument that grew directly out of my PhD research in AI and Music. For more information check out the website at netzxr.com . The instrument itself emerged from a research process focused on musical control, ergonomics, hand tracking, and expressive interaction in mixed reality. Netz was co-created through a longitudinal participatory design process with a professional…
This paper grew out of a practical problem in my XR musical instrument research: camera-based hand tracking works well until the fingers become occluded, which is exactly when musical interaction often needs the most precision. In this project, I explored a multimodal hand tracking pipeline that combines vision-based tracking from an XR headset with surface electromyography (sEMG) from a Myo…
This project was my first full XR musical instrument prototype study during the PhD. It became the foundation for much of the later work in Netz, multimodal hand tracking, and interactive machine learning for musical XR.
Computer-generated music can provide interesting insights into the structure of music and serve as inspiration for novice and professional composers alike. I investigate the use of the Transformer-XL neural network architecture for interactive co-creation of symbolic music in the style of pop piano. I present a modular system consisting of two software components: backend (music generation engine)…