#!/usr/bin/awk -f # # Make YouTube automatic subtitles more human readable # /^[0-9]+:[0-9]+:[0-9]+\.[0-9]+ -->/ { sub(/ align:start position:0%$/, "") if (caption && split(caption, lines, "\n") > 2) { print caption } ignore = 0 caption = $0 next } ignore { next } # skip extra empty lines /^ +$/ { next } # current caption contains garbage and will be repeated in the next # caption, ignore it caption && />/ { ignore = 1 caption = "" next } caption { caption = caption "\n" $0 next } { print } END { if (caption && split(caption, lines, "\n") > 2) { print caption } }