Find a recorded area, which has common crowd noise, with no distinguishable talking in it. If you can find a bland area like that, then it will be pretty easy. You just take those areas where the parents are talking and cut them out, and fill them with the bland crowd noise. Of course, if the pictures have cues that have sound, like a referee blowing a whistle, then this will take some advanced editing skills. Escentially, you can't remove what is there, you can only cut EVERYTHING out, and try to recreate the parts you wanted to keep. This type of of stuff is done in Audio Post production work all the time. I'll get a Jeep commercial, with this Jeep jumping through mud puddles, and driving over gravel, spinning it's tires and slamming on it's brakes. When it comes to me, it doesn't have any sound, I just use pre recorded sound effects and recreate everything. Try turning the volume down on your TV set and watch a TV commercial. That's how it comes into the audio post production house, hardly any of the audio is ever used from the actual video shoot.