Luca Comanducci

Dipartimento di Elettronica, Informazione e Bioingegneria

Building 20

via Ponzio 34/5, 20133 Milano, Italy

Phone: +39 02 2399 9640

I received the B.S. degree in Music Information Science from the University of Milan, Milan, Italy, in 2014, the M.S. degree in Computer Science and Engineering in 2018 and the Ph.D. (cum laude) in Information Engineering in 2022, both from Politecnico di Milano, Milan, Italy, where I am currently working as a temporary assistant professor (Ricercatore a Tempo Determinato di tipo A) as part of the Image and Sound Processing Lab within the Department of Electronics, Information and Bioengineering (DEIB). I have been research assistant for the INTERMUSIC project and I am currently part of the (REPERTORIUM) project. Since 2018 I have been teaching assistant in the M.Sc. in Music and Acoustic Engineering at Politecnico di Milano, for courses related to musical acoustics and creative programming. I have been part of the organizing committee of ISMIR 2023 as a member of the tech team.

My main research interests concern the application of generative models to music informatics, space-time audio signal problems, and the development of techniques for human-computer interaction in musical contexts.

Here you can find my full academic CV!

news

Jan 22, 2026	Starting 2026 with a good news, our paper Mitigating data replication in text-to-audio generative diffusion models through anti-memorization guidance co-written with Francisco Messina, Francesca Ronchini, Paolo Bestagini and Fabio Antonacci, has been accepted at the 2026 IEEE International Conference on Acoustics, Speech, and Signal Processing in Barcelona! In this paper, we explore how to use post-training techniques to avoid the generation of audio tracks memorized from the training set. here you can find the preprint of the paper https://arxiv.org/abs/2505.07615 and also the code is fully available at https://polimi-ispl.github.io/anti-memorization-tta/
Jul 09, 2025	I am super happy to announce that our paper Diffused Responsibility: Analyzing the Energy Consumption of Generative Text-to-Audio Diffusion Models (Riccardo Passoni, Francesca Ronchini, Romain Serizel, Fabio Antonacci) has been accepted at WASPAA 2025!!! In this paper, we apply Pareto analysis to examine the environmental impact of the use of TextToAudio models! here you can find the preprint of the paper https://arxiv.org/abs/2505.07615!
Dec 20, 2024	This year I have two papers accepted at ICASSP 2025 in India! MambaFoley: Foley Sound Generation using Selective State-Space Models Marco Furio Colombo, Francesca Ronchini, Luca Comanducci, and Fabio Antonacci arxiv link Towards HRTF Personalization using Denoising Diffusion Models, Juan Camilo Albarracín Sánchez , Luca Comanducci, Mirco Pezzoli, Fabio Antonacci arxiv link
Nov 21, 2024	Me and Francesca Ronchini are going to co-supervise a thesis in collaboration with INNER_SPACES the most adventurous avant-garde electronic event in Milan. Wenjia Luo, a student of the M.Sc. in Music And Acoustic Engineering is going to develop tools for enhancing acousmatic music performance!.
Oct 25, 2024	Francesca Ronchini is going to present our paper Synthetic Training Set Generation Using Text-to-Audio Models for Environmental Sound Classification at the DCASE workshop in Tokyo, Japan!