%0 Conference Proceedings
%T Improving Handwritten Music Recognition through Language Model Integration
%A Pau Torras
%A Arnau Baro
%A Alicia Fornes
%A Lei Kang
%B 4th International Workshop on Reading Music Systems (WoRMS2022)
%D 2022
%F Pau Torras2022
%O DAG; 600.121; 600.162; 602.230
%O exported from refbase (http://refbase.cvc.uab.es/show.php?record=3735), last updated on Thu, 27 Apr 2023 14:55:49 +0200
%X Handwritten Music Recognition, especially in the historical domain, is an inherently challenging endeavour; paper degradation artefacts and the ambiguous nature of handwriting make recognising such scores an error-prone process, even for the current state-of-the-art Sequence to Sequence models. In this work we propose a way of reducing the production of statistically implausible output sequences by fusing a Language Model into a recognition Sequence to Sequence model. The idea is leveraging visually-conditioned and context-conditioned output distributions in order to automatically find and correct any mistakes that would otherwise break context significantly. We have found this approach to improve recognition results to 25.15 SER (%) from a previous best of 31.79 SER (%) in the literature.
%K optical music recognition
%K historical sources
%K diversity
%K music theory
%K digital humanities
%U http://refbase.cvc.uab.es/files/TBF2022.pdf
%P 42-46