Publications

Latent Diffusion Bridges for Unsupervised Musical Audio Timbre Transfer
ICASSP 2025
Michele Mancusi, Yurii Halychanskyi, Kin Wai Cheuk, Eloi Moliner, Chieh-Hsin Lai, Stefan Uhlich, Junghyun Koo, Marco A. Martinez-Ramirez, Wei-Hsiang Liao, Giorgio Fabbro, Yuki Mitsufuji
Latent Diffusion Bridges for Unsupervised Musical Audio Timbre Transfer
COCOLA: Coherence-Oriented Contrastive Learning of Musical Audio Representations
ICASSP 2025
Michele Mancusi, Ruben Ciranni, Giorgio Mariani, Emilian Postolache, Giorgio Fabbro, Emanuele Rodolà, Luca Cosmo
COCOLA: Coherence-Oriented Contrastive Learning of Musical Audio Representations
High-Resolution Speech Restoration with Latent Diffusion Model
ICASSP 2025
Tushar Dhyani, Florian Lux, Michele Mancusi, Giorgio Fabbro, Fritz Hohl, Ngoc Thang Vu
High-Resolution Speech Restoration with Latent Diffusion Model
Multi-Source Diffusion Models for Simultaneous Music Generation and Separation
Oral (top 1%) @ ICLR 2024
Michele Mancusi, Giorgio Mariani, Irene Tallini, Emilian Postolache, Luca Cosmo, Emanuele Rodolà
Multi-Source Diffusion Models for Simultaneous Music Generation and Separation
Accelerating transformer inference for translation via parallel decoding
ACL 2023
Andrea Santilli, Silvio Severino, Emilian Postolache, Valentino Maiorca, Michele Mancusi, Riccardo Marin, Emanuele Rodolà
Accelerating transformer inference for translation via parallel decoding
Latent Autoregressive Source Separation
AAAI 2023
Michele Mancusi, Emilian Postolache, Giorgio Mariani, Andrea Santilli, Luca Cosmo, Emanuele Rodolà
Latent Autoregressive Source Separation
Towards the evaluation of marine acoustic biodiversity through data-driven audio source separation
I3DA 2023
Michele Mancusi, Nicola Zonca, Emanuele Rodolà, Silvia Zuffi
Towards the evaluation of marine acoustic biodiversity through data-driven audio source separation
Exploiting Music Source Separation For Singing Voice Detection
MLSP 2023
Michele Mancusi, Francesco Bonzi, Simone Del Deo, Pierfrancesco Melucci, Maria Stella Tavella, Loreto Parisi, Emanuele Rodolà
Exploiting Music Source Separation For Singing Voice Detection
Unsupervised Source Separation via Bayesian Inference in the Latent Domain
arXiv
Michele Mancusi, Emilian Postolache, Giorgio Mariani, Marco Fumero, Andrea Santilli, Luca Cosmo, Emanuele Rodolà
Unsupervised Source Separation via Bayesian Inference in the Latent Domain
Experimental device-independent certified randomness generation with an instrumental causal structure
Communications Physics
Iris Agresti, Davide Poderini, Leonardo Guerini, Michele Mancusi, Gonzalo Carvacho, Leandro Aolita, Daniel Cavalcanti, Rafael Chaves, Fabio Sciarrino
Experimental device-independent certified randomness generation with an instrumental causal structure