//SpeechSynthesis capability.speechSynthesis
change command speak(String text, Integer volume = null, String awsPollyVoiceName = null)
So it looks like in 2.2.6, the parameters of SpeechSynthesis changed but were not required and didn't send if null
Now in 2.2.9, the volume is sending whether selected or not (it is sending a null value).
ie. - old value being sent from RM using 'speak'
This is a test
new value being sent from RM using 'speak'
if volume is selected: This is a test, 35
if no volume: This is a test, null
Seeing the two ways of sending the speak aren't compatible, it doesn't work for people that use RM5.1 and User created app/driver (using the original speak).
So now I'll need to go in and change every app that uses speech to reflect the new way. (And so will every developer that uses 'speak'.)
This will take a few days. Hold tight. 
