QTextToSpeech::enqueue doesn't appear to work properly with winrt speech engine. If you call enqueue twice in a row, sometimes (but not always) the first text isn't spoken, just the second. Same code works fine with sapi engine.

Also, QTextToSpeech::aboutToSynthesize doesn't appear to work as documented for both winrt and sapi. In the docs, it says:

    void QTextToSpeech::aboutToSynthesize(qsizetype id)

The id is the value returned by a call to enqueue(), 

However, here's a list of ids returned by enqueue and then what is passed to a slot connected to aboutToSynthesize:

enqueue 0
aboutToSynthesize 0
enqueue 1
enqueue 2
aboutToSynthesize 0
enqueue 3
aboutToSynthesize 1
enqueue 4
aboutToSynthesize 2
aboutToSynthesize 0
enqueue 5
aboutToSynthesize 0
enqueue 6
aboutToSynthesize 0
enqueue 7
enqueue 8
enqueue 9
enqueue 10
aboutToSynthesize 3
aboutToSynthesize 4
aboutToSynthesize 5

If there's nothing queue, then the id passed to aboutToSynthesize appears to always be 0, not the id returned from enqueue which appears to increment sequentially.