- Investigating chord transitions and patterns/sequences in music is a crucial task for musicologists
- The automatic extraction of repetitive patterns based on different categorizations (e.g., composer, style, etc.) is a challenge that could benefit from employing interactive visualization
- Your target in this project is to develop or extend (CorpusVis) an interactive visual analysis prototype that supports the visual analysis of chord transitions based on sheet music data (MusicXML) that will be provided.
- You will implement different visualizations with a specific focus on the visual analysis of chord transitions in single compositions or sheet music collections.
- How can we use existing interactive visual techniques to support the exploration and identification of salient chord transition and relevant chord sequence patterns?
- What aspects are important when implementing an analysis prototype that supports the visual analysis and identification of chord transition and sequence patterns?
- How can we consider musicological knowledge to generalize the function of a chord from a specific key signature to its roman numeral function?
- There are four basic chord types: major, minor, augmented, and diminished chords. Identifying the function of a chord (numeral inside a scale) is a common task for musicologists. How can we use visualization to ease such analysis tasks?
- Implement a chord transition graph/network visualization or other suitable visual techniques
- Implement a ranking visualization for identified relevant chord sequence patterns
- Address performance and scalability issues enabling flexible loading of various datasets in the analysis prototype (dataset(s) will be provided)
- Identify recurring and salient chord sequence patterns within single compositions and sheet music collections based on metadata (composer, style, epoch, instrumentation, etc.)
- Implement interactive user functionality that enable music analysts to flexible analyze chord patterns in diverse sheet music collections
- Integrate Linking and Brushing between multiple components
- Interest in learning something about music data (e.g., key signatures, chords, music structure).
- You will use existing libraries to extract chords and apply algorithmic chord sequence analysis.
- Basic knowledge about information visualization and data mining.
- You will use a REST API with Python Flask or FastAPI
- Readiness to do creative research work independently
Scope / Duration / Start
- Scope: Bachelor / Master
- Project/Thesis Duration: 3 months / 3 months (Bachelor), 3 months / 6 months (Master)
- Start: Consider the project registration deadlines provided by the Department of Computer and Information Science (BA | MA)
- Bachelor / Master Project Guide
- M. B. Nardelli. Tonal harmony and the topology of dynamical score networks. Journal of Mathematics and Music, 2021: 1-15. DOI: 10.1080/17459737.2021.1969599
- Miller, J., Nicosia, V., & Sandler, M. (2021, July). Discovering Common Practice: Using Graph Theory to Compare Harmonic Sequences in Musical Audio Collections. In: Int. Conf. on Digital Libraries for Musicology (pp. 93-97). DOI: 10.1145/3469013.3469025
- Cabral, G., & Willey, R. Analyzing Harmonic Progressions with HarmIn: the Music of Antônio Carlos Jobim.Brazilian Symp. on Computer Music. 2007.
- Barthet, M., Plumbley, M., Kachkaev, A., Dykes, J., Wolff, D., Weyde, T. Big Chord Data Extraction and Mining. Proc. of Conf. on Interdisciplinary Musicology, 2014. URL
- M. Good. MusicXML for notation and analysis. The virtual score: representation, retrieval, restoration 12:113-124, 2001. https://www.musicxml.com/de/
- Cuthbert, M. S., & Ariza, C. music21: A toolkit for computer-aided musicology and symbolic music data. (2010). https://web.mit.edu/music21/