Earlier quoted context omitted.
If you've got a script then software like CMU Sphinx (open source) can use voice recognition to align the words of the script with the time of the soundtrack. Services like YouTube will even do that for you, if you can provide the bare minimum of a transcript. If even that sounds too hard, you can let YouTube's automatic subtitling produce an initial transcript - sometimes this can be surprisingly good, depending on…
Bluntly, when I make and release content for free I'm generally not very interested in paying for small boost in user numbers. Especially since for most videos, that 7-15% boost is maybe ten views.
And they don't seem to realise that it's not just deaf people that benefit from captions. There are a bunch of situations where I could really do with captions -- coffee shops; planes; late nights; etc etc.