Abstract
Optical Music Recognition is a field of research that investigates how to computationally decode music notation from images. Despite the efforts made so far, there are hardly any complete solutions to the problem. In this work, we study the use of neural networks that work in an end-to-end manner. This is achieved by using a neural model that combines the capabilities of convolutional neural networks, which work on the input image, and recurrent neural networks, which deal with the sequential nature of the problem. Thanks to the use of the the so-called Connectionist Temporal Classification loss function, these models can be directly trained from input images accompanied by their corresponding transcripts into music symbol sequences. We also present the Printed Music Scores dataset, containing more than 80,000 monodic single-staff real scores in common western notation, that is used to train and evaluate the neural approach. In our experiments, it is demonstrated that this formulation can be carried out successfully. Additionally, we study several considerations about the codification of the output musical sequences, the convergence and scalability of the neural models, as well as the ability of this approach to locate symbols in the input score.
Highlights
During the past few years, the availability of huge collections of digital scores has facilitated both the music professional practice and the amateur access to printed sources that were difficult to obtain in the past
We present the Printed Images of Music Staves (PrIMuS) dataset, containing more than 80,000 monodic single-staff real scores in common western notation, that is used to train and evaluate the neural approach
We study in this work a holistic approach to the task of retrieving the music symbols that appear in score images
Summary
During the past few years, the availability of huge collections of digital scores has facilitated both the music professional practice and the amateur access to printed sources that were difficult to obtain in the past. Some examples of these collections are the IMSLP (http://imslp.org) website with currently 425,000 classical music scores, or many different sites offering Real Book jazz lead sheets. The great possibilities that current music-based applications can offer are restricted to scores symbolically encoded An initial processing of the image is required This involves various steps of document analysis, not always strictly related to the musical domain. Results have reached values closer to the optimum over standard benchmarks by using DL [27,28]
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
