Re: new version of Enhanced phonetic reading (delayed descriptions)



If the BreakCommand is inaccurate for synthesizers included in NVDA (OneCore, SAPI5, eSpeak), this issue should be tracked. It has no sense to keep an half-working feature in NVDA and then say "I do not use this feature because it's not working" without reporting it formally.
Of course, this matter of facts has been signalled in the PR's discussion. But if the persons who have experienced this lack of accuracy could open a GitHub issue against NVDA with a full example (reproducible and measurable), this would really help fixing the BreakCommand issue, at least for synthesizers included in NVDA. If nothing can be done to fix it due to technical limitation or external blocking issue, maybe NVAccess will change its mind on this topic.

In any case, the add-on deserves its existence while the BreakCommand is inaccurate in NVDA or in other third-party add-on, which may be a very long time.



Join to automatically receive all group messages.