Issues related to unsupported fonts, languages, format
The article lists issues that can occur due to unsupported fonts, languages, or formats and solutions to address them.
Dots instead of letters in recognized PDFs
If a recognized document contains dots instead of text, it means that the OCR engine does not have the required fonts.
To import Microsoft fonts into your system, do as follows:
Copy the archive from the
PACKAGE_DIR/sources/folder on the MASTER server to the AGENT server.Execute the command on the AGENT server, specifying the location where the archive is to be copied.
tar -x -C / -f path-to-your-file/msttcore.tgz
OCR returns error message about language not supported
If, in an OCR response, you see <error><message>'Base' license category does not support the following language(s): KoreanHangul, ChineseTaiwan</message></error>, it means that your license type (Base in that case) doesn't support the mentioned languages. Only the English language is available for the Base license type.
To resolve the issue, make a request to the Workfusion Support Service Desk to update the license type.
OCR returns response message about checking failure of language argument
If, in an OCR response, you see message="Arguments check failed. Expected: !allowPredefinedLanguagesOnly", it means that OCR could not find the specified language among the predefined ones.
In the ETL OCR Settings, check the spelling of the language name. For instance, the Chinese language must be defined as _ChinesePRC_ or _ChineseTaiwan_.
For proper language names, refer to the official ABBYY OCR page.
OCR fails during processing due to unsupported format
If you get the following error: "The PDF file \\ABBYY\FineObjects\VirtualFiles$\StreamsFiles{2B9D045A-9113-4A20-AB08-DF7B5EE9D460} has unsupported format and cannot be opened", it may indicate there's an issue with the format of the input PDF document, its structure, encrypted containers inside it, internal components, or other.
You can avoid the problem by pre-processing—conversion to TIFF.
For OCR troubleshooting instructions, refer to the guide. For advanced OCR API, refer to the page.
Read also: