Music Informatics

Music Informatics is the scientific field that combines music with computer science, focusing on the study, analysis, retrieval and processing of musical information through computational methods. It utilizes digital audio, psychoacoustics, data mining, artificial intelligence and multimedia information technologies to understand and organize musical data. Its applications range from music categorization and retrieval to the development of music interaction systems, the study of popularity and the development of intelligent tools for the music industry and research.

The theoretical part of the course covers: the principles of acoustics and psychoacoustics of music, digital audio technology, techniques for extracting musical information features, methods for similarity and categorization of musical data, legal issues related to the use of musical information (copyright, fair use), as well as modern methodologies for knowledge mining using neural networks. User interfaces, visualization of musical data, queries with soft singing and the challenges of retrieving musical information in special environments (p2p networks, wireless networks, information flows) are also examined.

The laboratory part includes practical application using specialized software and open source tools for: analyzing and processing musical data, applying mining and categorization techniques, developing experimental musical information retrieval systems and implementing projects that combine music and computational methodologies. Students are invited to design and implement integrated music computing systems, which they will document and present at the end of the semester.

Upon successful completion of the course, the student will be able to:

• Analyze the structural elements of music computing applications
• Design and build music computing applications
• Understand the basic and critical characteristics of common actions in music computing
• Have knowledge of the tools and techniques of music computing and how they are used to achieve its goals.
• Understand and implement music computing methodologies with optimal programming solutions.
• Propose, design, implement, document and deliver a music computing system.

• Introduction: Acoustics - psychoacoustics of music, Digital audio technology, Extraction of musical information features, Similarity of musical data
• Legal issues of musical data: Intellectual property, Support for reasonable infringement, Basic actions of musical information retrieval and legislation
• Music data: Internet services, Sources of musical data tags, Data sets for researchers
• Similarity of musical information in acoustic data: Based on metadata, Based on content, Based on relevant social networking information & tags
• Knowledge mining from musical information: Recurring patterns, Clustering, Grouping
• User interfaces of musical information retrieval systems: The vocabulary problem, Queries with soft-singing, Visualization of musical data
• Retrieval of musical information in specialized environments: P2P networks,
• Wireless networks, Streams information
• Popularity of music tracks
• Clustering and classification of acoustic music data
• Deep neural networks and music

• Karydis, I. (2015). Introduction to retrieval and mining of musical information. Kallipos, Open Academic Editions. https://dx.doi.org/10.57713/kallipos-495
• Lotis, T., & Diamantopoulos, T. (2015). Music Informatics and Computer Music. Kallipos, Open Academic Editions. https://dx.doi.org/10.57713/kallipos-543
• Koutras, A., Alexandraki, C., Zarouchas, T., Zervas, P., & Chatziantoniou, P. (2023). Audio, speech and music processing and analysis. Kallipos, Open Academic Editions. https://dx.doi.org/10.57713/kallipos-347
• Δ. Πολίτη: "Γλώσσες και Διεπαφές στη Μουσική Πληροφορική", Εκδόσεις Κλειδάριθμος, Αθήνα 2008. Eudoxus code: 13630

• In-person interactive lectures
• In-person labs
• Optional project work

• Interactive whiteboard
• Slides with animation
• Computer software for digital music management and editing
• Design and development of computer music computing applications in Octave GNU, and Python

With optional assignment

• Grade = 60% assignment, 40% written exam
• Required pass condition: written exam grade >= 5/10

Without optional assignment

• Grade = 100% written exam
• Required pass condition: written exam grade >= 5/10