[suggestion] handwriting dataset

[Toon-nooT]

Hello, i 've created a new dataset with synthesized handwritten images.
68k images, 15 languages, rich metadata and quality filtered
It's opensource and maybe it proves useful for your future model trainings
https://huggingface.co/datasets/to-be/OpenHand-Synth