There are two major problems.
(1) there needs to be a blank line separating each of the entity types (before each "_____ is an entity type") and before each of the fact types
(2) we should try to preserve the indentation when doing a "copy image" and paste into a word processing application such as Word.
The first is critical. Even if you do the second, the first is still needed, but we could get by with just the first. If you don't preserve the indents, the blank lines (1) are even more needed. The blank line between the entity type and the facts in which it participates makes it even more confusing.
When I am grading student submissions and the verbalization has no indents (which is always), I find it very frustrating not to be able to find a particular entity. If the student got a constraint wrong in the ORM diagram I want to find and highlight it in the verbalization. If they had read it, they would have found the error in the model. I tell them that reading the verbalization is no less than what they would be asking and expecting the user domain experts to do.
When I document an ORM model, I try to cleanup the verbalization. That is almost impossible currently. In the attached .docx file you can see my attempt. I have also included the .orm file (oops, I guess I can't, too bad).