Voice overs for question wording (e.g., audio cati style)

[jarvis24]

[edited from support email]

I heard that you can program MediaLab to "read" the questions to the participant. So, if they have headphones, they can see and hear the text of the question before choosing a response. I have read the documentation and could not figure out how you would do this. (I think that backsounds won't work for this since it plays continuously and there seems to be more documentation about recording the respondent's voice, rather than the intereviewer's voice.) Would this require recording a separate sound file for each question, then "attaching" it to that question? Also, is there any mechanism in MediaLab to record these questions being read aloud, or would we have to create the mp3 file with another software?