datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
audio-dataset-flickr-soundnetdota_v1.5css-deepfake-datasetTraid_eval_dataAMSMB-line-transcription
Dataset Card
Dataset for line-level handwritten text recognition on medieval historical manuscripts, consisting of 3,369 lines (images of text lines with the associated transcription and metadata) from 100 digitized documents written by at least 80 different hands and spanning three centuries (from 1208 to 1499). This dataset is derived from the AMSMB dataset, which contains the full-page images of the digitized manuscripts and their associated transcriptions in the PageXML format.… See the full description on the dataset page: https://huggingface.co/datasets/BSC-CSSH/AMSMB-line-transcription.colmap_testsd_finetune_democsszengardenVVM-Tuning_Train_DataVVM-Bench