Alt Text
my bash script to clean up the formatting of text from books is also really useful for TTS purposes. the audio flows a lot more smoothly, and it in my most recent example it cut out 25 minutes of unnecessary pausing.
the final frontier will be to somehow exclude footnotes, however arbitrarily they are placed. but i'm kinda at a loss of how i might accomplish that. i assume it would have to be at the OCR level? and exclude font sizes that are too small? but that seems way above my skill level for now