You are viewing limited content. For full access, please sign in.

Discussion

Discussion

Cyrillic OCR

posted on August 1, 2023

Is there any known version of Laserfiche, present or future, self-hosted or cloud, that can do OCR in Cyrillic?

I've been evaluating Laserfiche Cloud and unless I've missed a setting somewhere, it's not (yet) able to accomplish this.

0 0
replied on August 1, 2023

You can change the default OCR language for the repository to one that uses Cyrillic, like Serbian, Bulgarian or Russian. The option is available in the "General" category under Repository Management.

Self-hosted components that can connect to cloud, like Import Agent or Quick Fields, can also be set to use one of these languages.

2 0
replied on August 2, 2023

Wow, Miruna, that's an even better answer than I was hoping for. I was hoping the capability was going to be in an upcoming version.

My follow-up question is no doubt harder to achieve technically without degrading accuracy, but after you so pleasantly surprised me, I have to at least ask: Is there a way to have the system OCR text correctly without knowing ahead of time whether a document is in a western European language or in a language like Ukrainian or Russian that uses a version of Cyrillic?

1 0
replied on August 2, 2023

Currently, there's no way for the system to auto-detect the language.

You could set up Quick Fields Agent to re-OCR documents and have your users tag them for re-processing as they run into them.

2 0
You are not allowed to follow up in this post.

Sign in to reply to this post.