We might be doing unecessary operations when Voicing/SpeechSynthesis is not enabled by stepping in this case.
The UtteranceQueue returns in step if not enabled. Maybe we can/should do the same for SpeechSynthesisAnnouncer. The UtteranceQueue used to step the SpeechSynthesisAnnouncer, so this actually seems like a safe thing to do after we moved the announcer.step call out of the UTteranceQueue in #71.
Perhaps in addition to this, we should add a listener that applies to chromium wake engine workaround to speak an empty string right when we enable, so that it doesn't interfere with that on chromium.
We might be doing unecessary operations when Voicing/SpeechSynthesis is not enabled by stepping in this case.
The UtteranceQueue returns in step if not enabled. Maybe we can/should do the same for SpeechSynthesisAnnouncer. The UtteranceQueue used to step the SpeechSynthesisAnnouncer, so this actually seems like a safe thing to do after we moved the announcer.step call out of the UTteranceQueue in #71.