In my work I also use a library with which I recognize text from PDF documents. However, depending on the quality of the document, the recognized words may contain spelling errors. I will give examples: instead of "RETURN", I get "RETОRN" instead of "either", I get "elther" instead of "copy", I get "copu" instead of "payment method", I get "paymentmethod" instead of "Accelerated Payment", I get "AcceleratedPayment" Note that these words are often recognized correctly, but on documents of poor quality, after recognition, there may be errors in the words. I have given these words purely as an example, spelling errors can occur in any words. However, I would like to correct such words with errors. I would like to draw your attention to the fact that I am not asking for a ready-made solution, I just want you to share with me good practices, links, so that I can solve this problem Continue reading...
