AI Audio Pronunciation Not Consistent

Hello,

I am trying to use AI text-to-speech audio in a Storyline course. There is one word used frequently throughout the course that the AI is not consistent on. The word is MEWP and should be pronounced as one word. However, the AI will sometimes pronounce it as mew-pee. 

Is there a way to guarantee the word is pronounced the same every time audio is generated? 

Thank you!