This paper summarizes and analyzes the use of Artificial Intelligence at the Centro di Sonologia Computazionale (CSC), located in the Department of Information Engineering (DEI) at Padua University. After a brief introduction to the work at CSC over the last seventy years, the focus will be on the two main areas studied today: music preservation and generation. To delve deeper into the CSC research, the following is a description of the key projects in these areas: an IEEE standard based on Convolutional Neural Networks (CNN) for magnetic tape preservation; a model to translate old music programming languages into more modern ones; a study on the collaboration between human composers and AI in generating new music that encompasses different fields such as music, philosophy, and computer science; and finally, a fine-tuned version of MusicGen for creating music related to taste and smell sensations.
Questa tesi riassume e analizza l’utilizzo dell’Intelligenza artificiale presso il Centro di Sonologia Computazionale (CSC), laboratorio del Dipartimento di Ingegneria dell’Informazione dell’Università di Padova. Dopo una breve introduzione sul lavoro svolto al CSC negli ultimi settant’anni, seguirà un focus su quelle che oggi sono le due principali aree di ricerca: la preservazione e la composizione musicale. Per approfondire questi due ambiti, i successivi capitoli tratteranno quattro esperimenti particolarmente rappresentativi: uno standard IEEE basato sulle Reti Neurali Convoluzionali (CNN) per la preservazione di nastri magnetici; un modello per tradurre in linguaggi di programmazione musicale moderni quelli non più utilizzati; uno studio che copre diversi campi, quali musica, filosofia e informatica, sulla collaborazione tra compositori umani e IA per creare nuova musica; e infine, una versione fine-tuned di MusicGen per generare musica in base alle sensazioni gustative e olfattive.
Computational Co-creativity at Centro di Sonologia Computazionale
GIURIATI, SARA
2024/2025
Abstract
This paper summarizes and analyzes the use of Artificial Intelligence at the Centro di Sonologia Computazionale (CSC), located in the Department of Information Engineering (DEI) at Padua University. After a brief introduction to the work at CSC over the last seventy years, the focus will be on the two main areas studied today: music preservation and generation. To delve deeper into the CSC research, the following is a description of the key projects in these areas: an IEEE standard based on Convolutional Neural Networks (CNN) for magnetic tape preservation; a model to translate old music programming languages into more modern ones; a study on the collaboration between human composers and AI in generating new music that encompasses different fields such as music, philosophy, and computer science; and finally, a fine-tuned version of MusicGen for creating music related to taste and smell sensations.| File | Dimensione | Formato | |
|---|---|---|---|
|
Giuriati_Sara.pdf
Accesso riservato
Dimensione
8.66 MB
Formato
Adobe PDF
|
8.66 MB | Adobe PDF |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/92501