Get blocks that appear in most sequences after multiple sequence alignment , for not biological text

Question

I have have many regular (non biological) texts. I would like to do multiple sequence alignment on them, and get all blocks that appear in more than x% of the texts, in the corr