Returning text from skill verse spoken text

Hi,

I’m looking for any information on how one can return plain text to a websocket call from a skill verse the spoken dialog of the skill. Often times I have to format the text differently so that it is pronounced properly, but I’d like to have it returned in the proper format… for example… the word:

“OK”

I have found I have to split it as “O K” in the skill; however I’d like to return it as “OK”. Any suggestions on what I can use? I’m sure I’m missing something.

Thank you,
Jay

Are you interested in the mycroft reply to the human’s spoken command?

Yeah. Not the phonetic reply Mycroft says, not the English text of the phonetic reply.