Swedish OCR
Archival handwriting, blackletter newspapers and modern EU documents.
v1 · 3 tasks
A task pairs a dataset with an image input: a cropped text line or a whole document page. Each task below has equal weight in the benchmark score.
View leaderboard →-
Swedish archival handwriting
Swedish archival text lines from the out-of-domain HTR evaluation release of Riksarkivet (the Swedish National Archives). The benchmark balances its sample across 19 archive and document source files.
Cropped text lineThe model reads an image cropped to one line of text. 1,000 images.
-
Swedish newspapers (Fraktur)
Swedish newspaper text in 19th-century blackletter type. Språkbanken produced the transcriptions, and the Swedish National Archives converted the material into line images.
Cropped text lineThe model reads an image cropped to one line of text. 1,000 images.
-
EU Official Journal · Swedish
Swedish pages from the 2023 Official Journal of the European Union, released in OJ4OCRMT. The benchmark uses the regular-page test subset, rendered at 300 dpi.
Whole document pageThe model reads an entire page and returns its text in reading order. 100 images.