Thanks to Cagri Dogan and وفيق طاهر for translations into Turkish and Arabic!
I updated the release, so translations are already included:
Antonio, thank you for such a detailed analysis and a detailed description of your actions.
> 4.6. When I'm in a text in English, I change my default synthesizer for English with SwitchSynth Add-On.
> 4.8. Select a word in the English text and translate and command NVDA+Y, followed by the letter D.
> 4.9. The QuickDictionary switches the voice for the profile 1 in Portuguese will speak when the translation, but does not automatically return to the English language that is currently configured by default in NVDA.
The fact is that the add-on now remembers as the default synthesizer the one that was configured at the time of launching NVDA.
I will change this behavior so that the add-on will remember the synthesizer, which is installed just before the pronunciation of the dictionary article.
And one more point - switching to the default synthesizer occurs only at the end of all spoken text. If the speech is interrupted - switching will not happen.
Unfortunately, I still don't understand how to technically implement the forced switching of the synthesizer when interrupting speech.
I continue to study this issue, I hope I will be able to solve it in the near future.
Good luck to all!