Using the Advanced OCR interface, you can optionally configure the OCR engine to look for characters that are unique to specific languages/character sets. For instance, if you have a collection of Korean documents, you will want to generate OCR that is optimized for extracting Korean characters. To do that, you'd check the "Korean" box in this interface prior to clicking Run.
Generally, you can select more than one language at a time. However, using more than one will increase the amount of time it takes to generate the OCR.
You can only optimize for one CJK language at a time (Chinese Simplified, Chinese Traditional, Japanese, Korean).