Exclusion of italicized text during voice generation.

[Bubareh]

This is a great extension, but there are some things missing in any TTS module. For example, there is a ban on reading italicized text out loud. Characters often describe their actions in a format such as action. Not always want to hear it. It would be cool to add an option to skip this text and have the action read aloud when the 'show text' option is enabled. Even better if the actions were displayed as text between audio clips of the message.