Thank you for the detailed feedback! Regarding "Speak Selected" and Firefox Reader View, these features use the browser's internal "Web Speech API" which generates audio outside of the standard HTML5 media elements (like those used by YouTube). Because this audio is handled by the browser/OS level directly rather than through standard media tags, it is currently impossible for my extension to intercept or control that specific audio stream (unless they add an Intercept API for that). I apologize for the limitation, it is a restriction of how Firefox handles these accessibility tools :(