I have some text written in Valser German, which is a Swiss dialect of German. This means that many words are spelled differently from regular German, ex. hellig = heilig, etc.
The text is printed using Gothic characters. I can't simply use Old German to recognise the page text, because (I assume) the built-in dictionary "corrects" the spelling of words into German. My earlier example mentioned means that when I recognise the page text, "heilig" gets turned into "hellig," which I don't want.
I tried to create a custom OCR language using Old German as the base language, and for dictionary I chose "None." However, this results in the text not being recognised correctly at all and gives mostly garbage. This doesn't seem to change if I select other options under Dictionary.
How can I recognise a page of text written in Gothic that is close to Old German, but has spelling variations? And how can I prevent the output from being spell-checked against the built-in German dictionary?
Thanks!
コメント
1件のコメント
Hi Justin,
I created a support ticket based on your question. A customer support representative will get back to you shortly.
サインインしてコメントを残してください。