I've created audio for every spoken number in English (up to a trillion, at least). I'm trying to string them together so that VoiceAttack is capable of reading off any number using these sound files. Unfortunately, whenever I string them together using the "Wait until sound completes before continuing command" option, there's a rather large delay between each sound, which causes it to become disjointed and unintelligible. Is it possible to reduce this delay between commands in order to make it sound fluid? I've attached a copy of what it sounds like currently, and what the target goal is to sound like (I'm certain the end result in VA won't be as clean as the target, but that is acceptable).