It will be nice to have support for the speech dispatcher protocol used by Linux (and MacOS?) for system text-to-speech. Windows seems to have their own API see [here](https://learn.microsoft.com/en-us/previous-versions/windows/desktop/ms720163(v=vs.85)) although it looks deprecated.
