Safetensors
t5
Edit model card

Swedish OCR Correction

This model is an updated version of https://huggingface.co/viklofg/swedish-ocr-correction

The model has been trained to correct OCR predictions by Abbyy, Tesseract, and a combination of those on newspaper from 1818-2018 (see A Two-OCR Engine Method for Digitized Swedish Newspapers ).

Please check the original model for more information.

This new model has been trained much longer and manages to outperform the previous one using the same train-test split.

Model CER WER
Original OCR 3.01 13.23
viklofg 1.92 7.41
KBLab 1.57 6.23
Downloads last month
15
Safetensors
Model size
300M params
Tensor type
F32
·
Inference API
Unable to determine this model's library. Check the docs .

Space using KBLab/swedish-ocr-correction 1