GAN-based Content-Conditioned Generation of Handwritten Musical Symbols
Gerard Asbert, Pau Torras, Lei Kang, Alicia Fornés, Josep Lladós · 2025
Read GAN-based Content-Conditioned Generation of Handwritten Musical Symbols by Gerard Asbert, Pau Torras, Lei Kang, Alicia Fornés, Josep Lladós (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2510-17869.txt.
Description
Gerard Asbert, with four co-authors, generates fake handwritten music symbols to help teach computers to read old scores. Real labeled examples of handwritten historic music are scarce, which limits optical music reading. The team trains a generative network at the level of single symbols, then builds full scores with engraving software.
Judged by eye, the made-up symbols look close to real handwriting. That realism marks progress toward synthetic scores that training sets can use. Readers interested in music archives get a route to more training data without years of hand labeling by experts.
Catalog record
| CREATOR | Gerard Asbert, Pau Torras, Lei Kang, Alicia Fornés, Josep Lladós |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 29 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2510-17869