Enriching Music Libraries [Self Funded Students Only]

Music is an important phenomenon in human civilization, about which we understand surprisingly little. The information contained in musical scores is the central resource for musicological research, performers, composers, and librarians.

Recently, large-scale libraries of digitised sheet music have appeared, notably the Petrucci Music Library. However, the information in such collections (images of sheet music) is not directly amenable to meaningful computational analysis. Although collections of symbolic music do exist, e.g. MuseData, Mutopia, these are on a much smaller scale. In the case of printed books, currently being digitised on an industrial scale, and for which OCR is a solved problem, analogous high-impact initiatives like Google Books, OpenLibrary, HathiTrust have developed sophisticated methods for searching, browsing, analysis, and retrieval of content, greatly enriching the collections. In contrast, for digital music score OCR is not a solved problem, currently relying on metadata supplied by humans as means of search and navigation.

This interdisciplinary project will apply cutting edge machine learning techniques to enrich sheet music collections with medium-level annotations, facilitating next-generation ways to interact with large sheet music collections, such as searching for patterns, statistical analysis; comparison and data fusion with audio recordings, etc. This breaks new ground in musicological research and will revolutionise music education and understanding of music, due to the scale on which annotations will be available.

The proposed research has many facets. One or more of the following areas will be addressed, depending on the student’s background: (1) facilitating search and retrieval by medium-level features, e.g. by timbre, texture, harmony, dynamic profile; (2) improving score-to-audio alignment to facilitate a new generation of audio-textual music libraries and improve the realism of orchestral audio synthesis; (3) improving the quality of score images (e.g. denoising) and the quality of computer-generated scores; (4) improving robustness of music OCR by aggregating evidence from a large corpus of scores.

The successful applicant will join the Visual Computing group in the School of Computer Science and Informatics, and its recently established Computational Music research sub-group. The Visual Computing group has a track record of excellent research, with significant research output in the recent REF return, spanning a wide range of topics in the fields of computer vision, computer graphics, geometric computing, computational music, and image and video processing. Working alongside this interdisciplinary team informally, at weekly seminars, etc., the student will have opportunities for research presentations, and for career growth in many of these areas.

Contact for information on the project: [Email Address Removed]

Keywords: Artificial Intelligence / Computer Vision / Data Science / Human Computer Interaction / Machine Learning

Academic criteria:  

A 2:1 Honours undergraduate degree or a master's degree, in computing or a related subject.  Applicants with appropriate professional experience are also considered. Degree-level mathematics (or equivalent) is required for research in some project areas. 

Applicants for whom English is not their first language must demonstrate proficiency by obtaining an IELTS score of at least 6.5 overall, with a minimum of 6.0 in each skills component. 

This application is open to students worldwide. 

How to apply: 

Please contact the supervisors of the project prior to submitting your application to discuss and develop an individual research proposal that builds on the information provided in this advert. Once you have developed the proposal with support from the supervisors, please submit your application following the instructions provided below 

This project is accepting applications all year round, for self-funded candidates via https://www.cardiff.ac.uk/study/postgraduate/research/programmes/programme/computer-science-and-informatics  

In order to be considered candidates must submit the following information:  

• Supporting statement  
• CV  
• In the ‘Research Proposal’ section of the application enter the name of the project you are applying to and upload your Individual research proposal, as mentioned above in BOLD 
• Qualification certificates and Transcripts 
• Proof of Funding. For example, a letter of intent from your sponsor or confirmation of self-funded status (In the funding field of your application, insert Self-Funded) 
• References x 2  
• Proof of English language (if applicable) 

If you have any questions or need more information, please contact [Email Address Removed]