# Korean OCR

An ultra-lightweight Korean text line recognition model that supports the recognition of Korean and numeric characters, with an average accuracy of 60.21%.

TrOCR is a Korean image-to-text model based on a vision encoder-decoder architecture, using DeiT as the image encoder and RoBERTa as the text decoder.

An OCR model supporting Korean initial sound recognition, using an improved tokenizer to address the traditional TrOCR's shortcomings in Korean initial sound recognition

This is a proof-of-concept model for Korean text recognition, trained on the TrOCR architecture, supporting Korean text extraction from images.