[Code](https://github.com/PRamoneda/pdf-difficulty) | [Paper](https://arxiv.org/pdf/2309.16287.pdf) | [Dataset](https://zenodo.com/record/8126801) | [Demo](https://musiccritic.upf.edu/pdf_difficulty/) 

## Paper
P. Ramoneda, J. J. Valero-Mas, D. Jeong & X. Serra, "Predicting performance difficulty from piano sheet music images," in Proceedings of the 24th International Society for Music Information Retrieval Conference, Milan, Italy, 2023.

## Model Description

The language model pre-trained in all the pieces of ISMLP for piano with the FC-Embedding layer described in the paper with the multi-task classifier.

---
license: agpl-3.0
---