Music · Artificial intelligence · Interaction design
David Dalmazzo
Researcher in music and artificial intelligence, interaction designer, and musician. Marie Skłodowska-Curie Postdoctoral Fellow at the Music Technology Group, Universitat Pompeu Fabra, leading ANIMA — AI tools for composing beyond twelve tones.
Music & AI
Transformer models of harmony (The Chordinator, ChromaFlow), psychoacoustic models of microtonal chord space, and machine learning for musical gesture recognition. PyTorch, TensorFlow, scikit-learn.
Interaction design
Interactive installations and real-time visuals built with C++ (openFrameworks, Cinder), Processing and Unity3D for museums, expos and theatre.
Music
Responsive sound and electronic music composed in Max/MSP. Jazz guitar player.
Teaching
Workshops and courses on programming and interaction design — Processing, Python, Max — at IAAC, Laboral, Elisava, Sapienza and more, for over fifteen years.
Research
All research →ANIMA — AI-based Interactive Microtonal Compositional Assistant
Marie Skłodowska-Curie Postdoctoral Fellowship · Music Technology Group, Universitat Pompeu Fabra, Barcelona · September 2025 – 2027
AI tools for composing beyond twelve tones: a psychoacoustic model of microtonal chord space and language models of 31- and 53-TET harmony. MSCA fellowship at the MTG, UPF.
Modelling harmony with transformers — MUSAiC
Postdoctoral researcher · Speech, Music and Hearing, KTH Royal Institute of Technology, Stockholm · 2022 – 2025
Three years at KTH in the ERC project MUSAiC (Bob L. T. Sturm): transformer models of chord progressions — The Chordinator, ChromaFlow — and studies of AI-generated music.
Musical gesture recognition — TELMI
PhD · Machine Learning Lab, Music Technology Group, Universitat Pompeu Fabra · 2016 – 2020
PhD at the MTG: machine and deep learning to recognise bowing, fingering and expertise from IMU, motion-capture and EMG sensors, within the H2020 project TELMI.
Selected publications
All publications →- 2026
Modeling 53-TET chord progressions with a microtonal GPT-2
AIMC 2026 — 7th Conference on AI Music Creativity, Berlin
- 2026
Transactions of the International Society for Music Information Retrieval
- 2026
ChromaFlow: modeling and generating harmonic progressions with a transformer and voicing encoding
MML 2024 @ ECML/PKDD — Communications in Computer and Information Science, Springer
- 2025
A computer application to explore 53-tone equal temperament harmonies through modal interchange
NIME 2025, Canberra, Australia
