We use tone to mark questions. If you listen closely, the TTS detects the question mark and it does raise its town on the last word. In reality, the tone change is more prominent. I can't describe it in writing. Maybe this will help you: https://en.wikipedia.org/wiki/Romanian_phonology#Intonation
Fair warning: to me the explanation in that article seems wrong. In the first example, it suggests that only the last word is different from a declarative sentence, but I'm sure a change is present throughout the whole sentence. I'd recommend finding recordings of actual speakers.