PhD position at the MTG to do research on automatic discovery of musical patterns
The Music Technology Group of the Universitat Pompeu Fabra in Barcelona opens a funded PhD position to do research on signal processing and machine learning methods to support musical exploration.
Supervisor: Xavier Serra
Application deadline: October 27th 2020
Starting date of PhD: Fall 2021
A properly created music corpus composed of audio recordings plus accompanying metadata can be used to understand and capture the essence of a particular musical style. Based on this assumption amd bulding on the corpora already collected in Dunya, the goal of this PhD position is to develop unsupervised approaches with which to discover the relevant melodic patterns that characterize different musical entities (e.g. artist, composer, form, style, ..). We will need to study different melodic representations that can be obtained from the continuous time series resulting from the audio analysis and explore different strategies for temporal segmentation. We will also need to consider different similarity measures and characterizations of the melodic patterns and focus on musically relevant ways to identify and characterize those patterns. Given the computational complexity of the problem, we will have to study optimization strategies for making the work feasible.
This PhD thesis will be done within the Audio Signal Processing Lab of the MTG an in the context of the Musical AI project. The thesis should contribute to the development of new audio processing, symbolic score processing, and/or machine learning methodologies, and they should be based on strong music knowledge and practical objectives.
The PhD position will be funded by the Spanish Ministry of Science and Innovation and the application should be done online through this link, specifying the Musical AI project (Reference: PID2019-111403GB-I00).
What we are looking for:
- A strong background in Computer Science (mainly Audio Signal Processing, Statistical Analysis, and Machine Learning).
- A strong background in music.
- A completed (by Sept. 2021) master’s degree in computer science or a related field (preferably with a thesis that documents your existing expertise in music technology).
- Someone passionate about research and music that shares our values.
The succesful candidate will enroll to the PhD program of the Department of Information and Communication Technologies of the UPF.
For any further information you can contact directly Xavier Serra.