The decision as to whether a doctorate makes sense or not depends on many different factors, and there is more than one way of approaching a doctorate. On this website, I have compiled information that may help in making an informed decision. Please note that this website is not intended to be comprehensive or balanced. It reflects and relates to my personal point of view and considers aspects that are particularly relevant for a doctorate in my research group in the field of Music Information Retrieval (MIR).
It is worth thinking about these aspects carefully before starting a doctorate. After making an informed decision, I hope you will perceive your doctorate as a challenging, fascinating, fulfilling, and formative phase (typically three to five years) of your life. For questions, feedback, and suggestions, please do not hesitate to contact me.
As mentioned in the QZB guide (a very informative document as I think), the question "Do I want to do a doctorate?" is more complex than you might think at first glance. Maybe the following aspects and more specific questions are helpful in approaching this question.
Before accepting an offer for a possible PhD position, look for a personal conversation with your potential supervisor and professor to clarify the key aspects of your doctorate. The following questions could play a role here:
In order to better understand the result of such a conversation, it is helpful to think of alternatives and to talk about them with independent people.
Common goals for doing a doctorate are advancing the knowledge in a specific research area and acquiring the skills for becoming an independent researcher and problem solver. As the above questions indicate, doing a doctorate involves much more than achieving novel scientific results (which is hard enough). Doing a doctorate also involves to
Concluding this section, I would like to emphasize that a doctorate does not require superhuman abilities. You will enjoy your doctorate
The revolution in music distribution, storage, and consumption has fueled tremendous interest in developing techniques and tools for organizing, analyzing, retrieving, and presenting music-related data. As a result, the field known as Music Information Retrieval (MIR) has matured into an independent research area related to many different disciplines, including signal processing, machine learning, information retrieval, musicology, and the digital humanities. This diversity opens up many opportunities for challenging, interdisciplinary, and fascinating research projects at the intersection of engineering, computer science, data science, artificial intelligence and humanities. In our research group, we aim to apply, understand, and further develop technologies to substantially advance the state of the art in MIR. Furthermore, we also aim at aspects of data and model understanding, cross-disciplinary applications, science communication, and education. In this way, our overachieving mission is to approach learning from different perspectives in the context of challenging MIR applications.
First, we apply and develop machine learning techniques to learn from training examples with the aim to make accurate predictions for previously unseen data.
Second, by learning from the experience of traditional engineering approaches, our objective is to understand better existing and to build more interpretable systems.
Third, learning with and from domain experts about musical works and their recorded performances, we explore the potential of computational tools by considering complex music scenarios of musicological relevance.
Given the complexity and diversity of music, MIR research has to account for various aspects such as the genre, instrumentation, musical form, melodic and harmonic properties, dynamics, tempo, rhythm, and timbre, to name a few. Furthermore, being an applied field of research, MIR research often proceeds in research cycles starting with a concrete application and continuing with task formulation, mathematical modeling, development of algorithmic approaches, implementation, data preparation, evaluation, experimentation, and reflecting on all steps in the context of the given application. The close interlocking of the various aspects of MIR research is both a curse and a blessing. On the one hand, one can feel overwhelmed by the variety and complexity of MIR research questions, and it is often unclear where to start to advance research. On the other hand, one can contribute to MIR research in many ways and achieve substantial progress when concentrating on certain aspects.
Our research group is anchored in the International Audio Laboratories Erlangen. We look at MIR research with a focus on the processing and analysis of music signals while assuming a more technically oriented perspective. We particularly welcome doctoral candidates who
In case you want to find out more about our research group and interets, you may find the following links useful:
Most doctoral students in my group work as research associates employed by the Friedrich-Alexander-Universität Erlangen-Nürnberg and are a member of the International Audio Laboratories Erlangen. They formally work in the public sector of the federal state of Bavaria, and as for most doctoral candidates in engineering, they obtain a 100% position at the salary level TV-L E 13. I try to provide such a position for every doctoral student throughout the doctorate. However, such positions are rare and difficult to obtain. Besides qualification positions assigned to a professorship, most of these positions come from third-party funding sources. For example, I have some project positions financed by the German Research Foundation (DFG), where the project's content is usually directly linked to the doctorate. Depending on the position and project scope, a limited amount of teaching, including the supervision of students and student research projects, is part of the official tasks. This type of position is ideal for a doctorate and an academic career.
The International Audio Laboratories Erlangen (AudioLabs) are a joint institution of Fraunhofer IIS and Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU). Thanks to a close connection between FAU and Fraunhofer IIS, we can offer different models for doing a doctorate with either a more academic or industry-related orientation. In the following, we introduce a specific doctorate path called Doctorate with Fraunhofer as supported by the Fraunhofer Society.
The Fraunhofer society attaches great importance to supporting its employees in their individual career development. One aim is to allow young scientists to do a doctorate (formally supervised by a university professor) during their time at Fraunhofer. The requirements for doctoral candidates at Fraunhofer are described in the Code of Conduct. This document covers topics including the specification of a doctoral project, the organization of doctoral supervision, and responsibilities of the people and institutions involved.
Doing a more traditional doctorate working at a university, the professor often takes over the role of direct manager, scientific supervisor, mentor, and first reviewer. In contrast, in the Fraunhofer model, these roles are assumed by a team of several people, which forms the thesis advisory committee (TAC). The responsibilities may be summarized as follows (note that the lists are not exhaustive, and the division into roles is not defined strictly):
For this model to work, the candidate needs good organizational and communication skills to get support and feedback from the different people involved. Furthermore, in practice, creating suitable synergies in content and structure between the project work and the doctorate will be essential. A trusting relationship between the candidate and the committee is necessary to reconcile the many (sometimes conflicting) aspects.
QualitätsZirkel Promotion (QZP): Doing your doctorate. Making conscious decisions and getting off to a good start, 1st edition, 2020
QualitätsZirkel Promotion (QZP): Shaping a Doctorate Together – Guidelines for Doctoral Candidates, 3rd edition, 2018
QualitätsZirkel Promotion (QZP): Shaping a Doctorate Together – Guidelines for Supervisors, 3rd edition, 2018
Zeit Campus: Ratgeber Promotion, 2023
Fraunhofer: Doctorate with Fraunhofer – Code of Conduct, 2019
DFG: Guidelines for Safeguarding Good Research Practice. Code of Conduct. [PDF]
Thesis e.V.: Das deutschlandweite Netzwerk für Promovierende und Promovierte
DSZ: Deutsches Stiftungszentrum, Dienstleistungszentrum des Stifterverbandes für Stifter und Stiftungen
I would like to express my gratitude to my former and current students, collaborators, and colleagues for their feedback and the many discussions on this topic. In particular, I thank Andreas Brendel for his valuable suggestions.