{"authors": [{"name": "Rabus, Achim" , "affiliation" : "University of Freiburg"},{"name": "Thompson, Walker Riggs" , "affiliation" : "Heidelberg University"}, {"name": "Stökl Ben Ezra, Daniel" , "affiliation" : "École pratique des hautes études"}], "summary": "Generic HTR model suitable for transcribing a variety of Old Cyrillic script styles including uncial and semi-uncial (11th-16th c.)", "description" : "Training data consist of parts of the Russian Church Slavonic Great Reading Menology (16th century), Old Church Slavonic Codex Suprasliensis (11th century), and the 11th century manuscript of the Catecheses of Cyril of Jerusalem. This is a generic model suitable for transcribing a variety of Old Cyrillic script styles including uncial and semi-uncial. The original train set was prepared in Transkribus, whence it was exported and re-used to train this model. It is possible that the export caused some distortions of baselines or line masks, or corruptions in the data, which may have inflated CER (despite manual cleansing prior to Kraken training).", "accuracy": 94.3, "license": "CC BY-SA 2.0", "script": ["Cyr"], "name": "Cyr02full_best.mlmodel", "graphemes": [" ", "-", ".", ":", "m", "z", ",", ";", "E", "M", "O", "Y", "Z", "a", "e", "o", "\u00a0", "\u00ad", "\u00b7", "\u02bc", "\u0302", "\u0305", "\u0306", "\u0307", "\u0308", "\u030f", "\u0311", "\u0335", "\u033e", "\u0404", "\u0405", "\u0406", "\u0407", "\u0410", "\u0411", "\u0412", "\u0413", "\u0414", "\u0415", "\u0416", "\u0417", "\u0418", "\u041a", "\u041b", "\u041c", "\u041d", "\u041e", "\u041f", "\u0420", "\u0421", "\u0422", "\u0423", "\u0425", "\u0426", "\u0427", "\u0428", "\u0429", "\u042a", "\u042b", "\u042c", "\u042e", "\u0430", "\u0431", "\u0432", "\u0433", "\u0434", "\u0435", "\u0436", "\u0437", "\u0438", "\u043a", "\u043b", "\u043c", "\u043d", "\u043e", "\u043f", "\u0440", "\u0441", "\u0442", "\u0443", "\u0444", "\u0445", "\u0446", "\u0447", "\u0448", "\u0449", "\u044a", "\u044b", "\u044c", "\u044e", "\u044f", "\u0454", "\u0455", "\u0456", "\u0457", "\u0460", "\u0461", "\u0462", "\u0463", "\u0464", "\u0465", "\u0466", "\u0467", "\u0469", "\u046b", "\u046d", "\u046f", "\u0471", "\u0472", "\u0473", "\u0475", "\u0477", "\u0479", "\u047e", "\u047f", "\u0481", "\u0483", "\u0484", "\u0485", "\u0486", "\u0487", "\u04c0", "\u1fc0", "\u2040", "\u2058", "\u2219", "\u223b", "\u2c55", "\u2de0", "\u2de1", "\u2de2", "\u2de3", "\u2de4", "\u2de5", "\u2de6", "\u2de7", "\u2de8", "\u2de9", "\u2dea", "\u2deb", "\u2dec", "\u2ded", "\u2dee", "\u2def", "\u2df0", "\u2df3", "\u2df6", "\u2df7", "\u2df9", "\u2dfa", "\u2dfb", "\u2dfc", "\u2dfd", "\u2e0c", "\u2e3a", "\ua641", "\ua64b", "\ua64c", "\ua64d", "\ua651", "\ua656", "\ua657", "\ua659", "\ua66f", "\ua674", "\ua675", "\ua67a", "\ua67c", "\ua67f", "\ue049", "!", "\"", "/", ">", "C", "I", "N", "T", "X", "l", "x", "\u00cc", "\u0301", "\u0312", "\u0313", "\u0314", "\u0315", "\u0343", "\u0358", "\u035b", "\u035e", "\u0368", "\u0387", "\u03aa", "!", "\"", "/", ">", "C", "I", "N", "T", "X", "l", "x", "\u00cc", "\u0301", "\u0312", "\u0313", "\u0314", "\u0315", "\u0343", "\u0358", "\u035b", "\u035e", "\u0368", "\u0387", "\u03aa", "\u03b4", "\u03bf", "\u03c1", "\u03c5", "\u03c7", "\u03ce", "\u0424", "\u047b", "\u1f41", "\u1fe6", "\u2010", "\u203b", "\u205d", "\u2c58", "\u2df1", "\u2df2", "\u2df4", "\u2e2f", "\ua640", "\ua644", "\ua645", "\ua64a", "\ua665", "\ua676", "\ua677", "\ua678", "\ue20d", "\ufe21"]}