{"authors": [{"name": "Benjamin Kiessling", "affiliation": "\u00c9cole Pratique des Hautes \u00c9tudes, Aoroc - CNRS PSL"}], "summary": "Printed Urdu Base Model Trained on the OpenITI Corpus", "description": "This is a text recognition model trained on the OpenITI dataset of printed\nArabic-script text available at [0] in its state of 2022-09-03. It encompasses\nUrdu (~11k lines) material in a variety of typefaces. The model has been\nobtained by fine-tuning the Arabic-script base model [1] on the purely Urdu\nsubset of the corpus.\n\nAs the model is trained on a variety of highly diverse typefaces it is mostly\nintended as a base model for fine-tuning more specific models from it. In line\nwith this it has not been extensively verified or optimized.\n\nThe ground truth was lightly normalized to NFD but is otherwise untouched. \n\n[0]: https://github.com/OpenITI/arabic_print_data.git\n[1]: 10.5281/zenodo.7050270\n", "accuracy": 95.87959051132202, "license": "CC0-1.0", "script": ["Arab"], "name": "urdu_best.mlmodel", "graphemes": [" ", "!", "\"", "'", "(", ")", "*", "+", ",", "-", ".", "/", "0", "1", "2", "3", "4", "5", "6", "7", "8", "9", ":", ";", "<", "=", ">", "?", "A", "B", "C", "D", "E", "F", "G", "H", "I", "J", "K", "L", "M", "N", "O", "P", "R", "S", "T", "U", "V", "W", "X", "[", "\\", "]", "_", "a", "b", "c", "d", "e", "f", "g", "h", "i", "j", "k", "l", "m", "n", "o", "p", "r", "s", "t", "u", "v", "w", "x", "y", "z", "{", "|", "}", "\u00ab", "\u00bb", "\u060c", "\u060e", "\u0610", "\u0611", "\u0612", "\u0613", "\u0614", "\u061b", "\u061f", "\u0621", "\u0627", "\u0628", "\u0629", "\u062a", "\u062b", "\u062c", "\u062d", "\u062e", "\u062f", "\u0630", "\u0631", "\u0632", "\u0633", "\u0634", "\u0635", "\u0636", "\u0637", "\u0638", "\u0639", "\u063a", "\u0640", "\u0641", "\u0642", "\u0643", "\u0644", "\u0645", "\u0646", "\u0647", "\u0648", "\u0649", "\u064a", "\u064b", "\u064c", "\u064d", "\u064e", "\u064f", "\u0650", "\u0651", "\u0652", "\u0653", "\u0654", "\u0655", "\u0656", "\u0657", "\u0660", "\u0661", "\u0662", "\u0663", "\u0664", "\u0665", "\u0666", "\u0667", "\u0668", "\u0669", "\u066a", "\u066d", "\u0670", "\u0671", "\u0674", "\u0679", "\u067e", "\u0686", "\u0688", "\u0691", "\u0698", "\u06a4", "\u06a9", "\u06ad", "\u06af", "\u06ba", "\u06be", "\u06c1", "\u06c3", "\u06cc", "\u06d2", "\u06d4", "\u06d5", "\u06db", "\u06f0", "\u06f1", "\u06f2", "\u06f3", "\u06f4", "\u06f5", "\u06f6", "\u06f7", "\u06f8", "\u06f9", "\u0763", "\u200f", "\u2018", "\u2019", "\u201c", "\u201d", "\u2217", "\u2606", "\ufd3e", "\ufd3f", "\ufdf2", "\ufdfa", "\ufe85"]}