Question: extract only one language content in a dual language document using aws textract

[SpAcY001]

I have a bilingual contract containing text in both English and [Native Language]. I want to Extract only the content written in English, and ensure no content from the [Native Language] is included in the output.

only extract and display the English text exactly as it appears in the document, the document contains table, these are scanned documents, is this possible?

[Belval]

Out of the box Amazon Textract does not support this use case. You could use Amazon Comprehend (or an open-source library) to detect the language and remove any layout block that is not English, leaving you only with the English version?