A data-driven model of tonal chord sequence complexity
- Submitting institution
-
Queen Mary University of London
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 486
- Type
- D - Journal article
- DOI
-
10.1109/TASLP.2017.2756443
- Title of journal
- IEEE/ACM Transactions on Audio, Speech, and Language Processing
- Article number
- -
- First page
- 1
- Volume
- 25
- Issue
- 11
- ISSN
- 2329-9290
- Open access status
- Compliant
- Month of publication
- September
- Year of publication
- 2017
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
3
- Research group(s)
-
-
- Citation count
- 3
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- First large-scale analysis of harmony in Western popular music. We built a dataset of 0.5 million chord sequences and showed how to model this with a combination of a deep recurrent neural network, a hidden Markov model, and prediction by partial matching. This has applications for computational creativity, generative composition and music recommendation: our model is able to predict complexity ratings of human listeners, and we also show its relevance for understanding musical preference. Work in collaboration with a leading Italian university (Politecnico di Milano) and led to the visiting PhD student Di Giorgi getting a job with Apple (https://uk.linkedin.com/in/brunodigiorgi).
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -