Prerequisite/Recommended prerequisite for participation in the module

Content, progress and pedagogy of the module

Explore the development and analysis of practical and automatic methods for making accessible information contained in abstract formats of music, such as symbolic (sheet music), or digital audio samples (MP3), i.e., all the information that currently requires experienced human to extract. These include various tasks in which one can analyze music signals for, e.g., inferring or identifying the artist and song playing in a noisy environment like a pub (c.f. Shazam), organizing a music collection by genres (e.g., blues and/or hip hop), mood (e.g., restful or excited), or use (e.g., relaxation or exercise), determining the instruments playing in a recording (e.g., guitar and gong), the recording type (e.g., live or studio), (un)recommending music (e.g., “if you like Gustav Winckler, then you will not like L.O.C.”), creating playlists (e.g., “suggest a mix of songs from my collection for my new girlfriend”), composing new music (e.g., “mash together this Gustav Winckler song and that L.O.C. song”), automatic mastering (e.g., “what changes do I need to make to my song to make it more Pop-sounding?”), and so on.

Learning objectives

Knowledge

Students who complete the module will gain knowledge as follows:

• Must be able to describe and distinguish between methods of content classification, retrieval and description for audio and music signals.
• Must be able to describe the structure of systems for audio or music classification, retrieval, and description.
• Must be able to distinguish between supervised and unsupervised learning, and how they are used in music information research.
• Must be able to identify and describe low-, mid- and high-level representations of sound and music, and how they are used in music information research.
• Must be able to summarize the importance and relevance of human perception for music information research.
• Must be able to summarize and distinguish the experimental designs and figure of merits to use in music information research.

Skills

Students who complete the module will gain skills as follows:

• Must be able to analyze and compare a variety of approaches to audio and music content classification, retrieval, and description.
• Must be able to explain and compare a variety of approaches to evaluating systems for audio and music content classification, retrieval, and description.
• Must be able to explain the concepts behind a complex integrated system for working with the contents of audio and/or music signals, e.g., a sound search engine, query-by-humming or –example, music identification through fingerprint comparison, speech-driven menu system, and so on.
• Must be able to analyze the approaches and algorithms applied in a piece of scientific literature in music information research, interpret the assumptions made, and relate them to the goals of the work.
• Must be able to implement and reproduce a piece of scientific research literature in music information research, interpret the results, and compare them with those of the original research.
• Must be able to explain and argue for all assumptions made in the re-implementation.

Competences

Students who complete the module will gain competences as follows:

• Must be able to evaluate and criticize within the format of a critical annotation a recent piece of scientific literature (journal articles and conference papers) related to music information research (identifying its relevance and the perspective of the authors, stating the scientific hypothesis, theory, approaches and solutions, assumptions made, and its conclusions and contributions).
• Must be able to discuss and evaluate a complex integrated system for working with the contents of audio and/or music signals, e.g., a sound or music search engine, query-by-humming or –example, music identification through fingerprinting, recommender systems, and so on.
• Must be able to choose and judge frameworks for music information retrieval in a variety of practical situations.

Type of instruction

Academically supervised student-governed problem oriented project work.

Exam

Exams