speech-dispatcher in flatpaks
With the Fedora flatpak of Firefox, I sometimes get the following notice on certain sites:
(i) You can’t use speech synthesis because the Speech Dispatcher library is missing. Learn more [Don't show again]
Also, org.kde.kmouth (WIP) doesn't appear to be functional because it wants to use the speech-dispatcher backend of QtSpeech, while we only have the flite backend.
speech-dispatcher includes a systemd unit file, so it appears to be a host service, in which case we wouldn't want the whole thing in a flatpak. However, clients should be using either libspeechd or python3-speechd.
If that all is indeed the case, then perhaps it makes sense to break things up somewhat so that the client components can be flatpak'ed (possibly in the runtime, at least on the KDE side, for QtSpeech)?
- libspeechd would need to be split out into a new speech-dispatcher-libs subpackage;
- python3-speechd's dependency on speech-dispatcher would need to be softened or conditionalized.
Does this make any sense?