[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: HELP: Debug gnome-speech hang

Hello Mario,

I've just tried orca/upstream 2.17.4 on gnome/sid (2.14.3), with
python-pyorbit/experimental.  AFAIK, gnome-speech didn't hang, but
orca is... very unstable.  Also this page[1] mentions that
synthesis-drivers can hang.

I've installed libgnome-speech3-dev to run test-speech: yes, it hangs
when entering 'all' to choose a locale.  If I put "1" without asking
"all", it finds no voices; if I put "all", it speaks voices specs and
then my CPU tries to burn itself...

Before digging (down...) the ORB, I rather suspect
festival_createSpeaker function defined in
gnome-speech-0.3.10/drivers/festival/festivalsynthesisdriver.c, which
is a callback for GNOME_Speech_SynthesisDriver_createSpeaker called in

Did you get the problem with python-pyorbit/sid too ?


[1]  http://live.gnome.org/Orca/GnomeSpeech

Mario Lang <mlang@debian.org> writes:

> Hi.
> For some to me totally unknown reason gnome-speech
> (in particular, the festival-synthesis-driver binary) hangs with
> 100% CPU load whenever I try to invoke a gnome-speech app
> on two machines of mine, both running Debian Sid (one is amd64, the otehr
> is i386).  This used to work, I know!
> For instance, I call "orca --setup" in a terminal, which
> hangs, since it apparently tries to use gnome-speech, which in turn,
> hangs...
> Can anyone please test if they can get gnome-speech to
> work on their debian machine, and report back with results?
> Besides, everything you can figure out to further pin this bug
> down would be highly welcome.
> Skimming the news entries for newer gnome-speech versions,
> I can not find anything that reads like being relevant.
> Anyone with some CORBA-fu?
> -- 
> Thanks,
>   Mario | Debian Developer <URL:http://debian.org/>
>   .''`. | Get my public key via finger mlang@db.debian.org
>  : :' : | 1024D/7FC1A0854909BCCDBE6C102DDFFC022A6B113E44
>  `. `'
>    `-      <URL:http://delysid.org/>  <URL:http://www.staff.tugraz.at/mlang/>

Boris Daix

Reply to: