Music Processing

Art der Arbeit:

Master Thesis

Betreuer:

Müller, Meinard
Lehrstuhl für Semantische Audiosignalverarbeitung (AudioLabs)
Telefon +49 9131 85-20504, E-Mail: meinard.mueller@audiolabs-erlangen.de

Beschreibung der Arbeit:

The digital revolution in music distribution and storage has simultaneously fueled tremendous interest in and attention to the ways that information technology can be applied to music. Many research efforts are directed towards the development of technologies that allow users to access and explore music in all its different facets. For example, audio fingerprinting techniques are nowadays integrated into commercial products that help users to organize their private music collections. Music processing techniques are used in extended audio players that highlight the current measures within sheet music while playing back a recording of a symphony. On demand, additional information about melodic and harmonic progressions or rhythm and tempo is automatically presented to the listener. Interactive music interfaces display structural parts of the current piece of music and allow users to directly jump to any key part such as the chorus section, the main musical theme, or a solo section without tedious fast-forwarding and rewinding. Furthermore, listeners are equipped with Google-like search engines that enable them to explore large music collections in various ways. For example, the user may create a query by specifying a certain note constellation, or some harmonic or rhythmic pattern by whistling a melody or tapping a rhythm, or simply by selecting a short passage from a CD recording; the system then provides the user with a ranked list of available music excerpts from the collection that are musically related to the query.

Within the wider areas of music processing, audio processing, and music information retrieval (MIR), we offer various topics for Master theses that are related to current research projects of the group. The general objective of a Master thesis is to contribute concepts, models, algorithms, implementations, and evaluations for tackling a specific music or audio processing problem.

Examples for such problems are:

• Content-Based Audio Retrieval

• Audio Decomposition and Source Separation

• Tempo, Beat, and Ryhtmic Analysis

• Audio Structure Analysis

• Harmonic Analysis

• Music Synchronization

Vorausgesetzte Vorlesungen bzw. Kenntnisse:

A master thesis in this area requires a very good mathematical background and some strong personal interest in music. In particular, a very good understanding of aspects from digital signal processing and/or information science are needed. It is assumed that the candidate is familiar with the topics of the lecture "Music Processing Analysis" (see also the book "Fundamentals of Music Processing" by Meinard Müller, Springer 2015). Furthermore, the candidate should have excellent programming skills in at least one programming language (e.g. C/C++, MATLAB, Python).

Weitere Informationen zur Arbeit:

https://www.audiolabs-erlangen.de/fau/professor/mueller

Schlagwörter:

AudioLabs audio music retrieval analysis processing

Bearbeitungszustand:

Die Arbeit ist noch offen.