Czech diacritics are part of the word
Czech uses marks above and below its letters that change which word it is rather than only how it sounds. Stripping them, as some older subtitle tooling and a few upload pipelines still do, produces text that is wrong rather than merely plain.
Open the exported file once and confirm the marks survived before you rely on a workflow. If they did not, the problem is almost always an encoding setting rather than the Czech text itself.

