An End-to-End Neural Network for Polyphonic Piano Music Transcription
- Submitting institution
-
Queen Mary University of London
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 435
- Type
- D - Journal article
- DOI
-
10.1109/TASLP.2016.2533858
- Title of journal
- IEEE/ACM Transactions on Audio, Speech, and Language Processing
- Article number
- -
- First page
- 927
- Volume
- 24
- Issue
- 5
- ISSN
- 2329-9290
- Open access status
- Out of scope for open access requirements
- Month of publication
- February
- Year of publication
- 2016
- 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
-
2
- Research group(s)
-
-
- Citation count
- 70
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- Musicians struggle to play songs "by ear", so there is considerable interest in automatic transcription of music. We were first to show the possibility of combining a language model as used in speech recognition with an acoustic model to achieve state-of-the-art transcription performance. This highly cited work is used as a benchmark for transcription systems. It led to PhD of Sigtia (2017), who now works for Apple (https://uk.linkedin.com/in/siddharth-sigtia-47b1464b). Our system was reimplemented and improved by Google Brain (http://ismir2018.ircam.fr/doc/pdfs/19_Paper.pdf) and Wang (https://www.mdpi.com/2076-3417/8/3/470); we are collaborating with Apple (mmauch@apple.com) to extend the work to instrument mixtures.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -