Re: [orca-list] interesting performance with speech-dispatcher and orca.

[Tomas Cerha <cerha brailcom org>]

Hermann napsal(a):
> So I stopped SD and redid the setup. Everything seems fine, till I restart 
> Orca on the desktop. Orca does not speak anymore, and it is impossible to 
> choose a synth voice, I get the word "menu" and can't choose anything. This 
> happens also when I choose Festival as a synth which has only one voice.
> So in the end everthing's broken, and I have no idea to get it back to 
> work, congratulation!

Hello Hermann,

I agree this is a problem.  I actually have the same problem with Gnome
Speech.  It doesn't work on my system and I got to the same situation as
you with the SD backend.  I had to manually edit my
~/.orca/user-settings.py to get back to work.

My problem with Gnome Speech may be because of some package versions
mismatch (I am using a mixture of Debian testing/unstable/experimental)
so I am far from saying this is a bug in Gnome Speech, but test-speech
fails with a segfault.  But this doesn't matter for what I want to
suggest here.

I think that Orca might be more resistant to a failure in the speech
subsystem on startup.  When Orca starts the preferences dialog, it tries
to detect the available speech systems.  The same detection might take
place on startup and if the configured speech system doesn't work, it
migh be possible to use the next available system.  Does that make sense?

Best regards, Tomas