I always build my menues in FCP and Export > QT Movie.
I always let DSP3 do my encoding.
If I have several menues all with the same audio, I Export > QT Movie video only for each menu, then I export the audio clip. If it is a looping sound, I export the shortest amount of the loop (or loops).
In DSP I Add menu... then I drop the .mov file for that menu into that menu... then I check the length of the menu with the full length of the .mov file for the menu... then I add the audio clip in as many multiple times as necessary to add up to the length of the video.
This way, DSP does not waste tons of filespace encoding the same piece of audio over and over again.
I have found, that if the sum of the lengths of the audio clips exceeds the length of the video for the menu, DSP jumps to the next event based on the video length anyway, so not a problem to be sloppy.