Speech recognition and text reinterpretation for clear and error-proof text

Not speech-to-text, but speech-to-idea-to-text

I am writing this article with the speech recognition integrated in Google Docs, called Voice typing, and then translated into English.

It has some very interesting features, some sort of special commands such as go to new line, write punctuation, indent text and so on.

Unfortunately most of these features seem to me to be missing in the language I am writing this article (Italian), but I hope that in the future they will also add this functionality, it could be very interesting in terms of minimizing the time in writing.

I would like to discover a software, implementable for example as a browser extension, that has speech recognition feature, but not only, so not as a normal speech recognition that literally transcribes what you are saying, but that reinterprets your ideas that it hear in clear text consistent with what you want to write avoiding grammatical errors, and allowing you to write quickly even without the appropriate means such as a keyboard for example when you are traveling, because in fact a microphone is enough, which practically any smartphone has.

Not speech-to-text, but speech-to-idea-to-text

If you know some software like this described, I would appreciate know it in the comments.

P.s. If this software exist, I think it shouldn’t be used for artistic things, like romances or poetries.