I think we all understand about not having enough time.
I spent a bit of time this afternoon making some changes. The files
from the software TRM that we need to make the library are in current/
Frameworks/Tube. I've also moved the model bits from Monet into
Frameworks/GnuSpeech. These parts have been committed.
I also have Monet building against these on Mac OS X. I'll commit
those changes in a bit, but I don't have a GnuStep system so it's
going to break building on that platform. Next up will be moving the
model parts of the little GnuSpeech app into the framework. Then I
need to take another look at the CoreAudio stuff I'm using, since
they've deprecated some of the API I use.
After that, well, I'll probably go back into hiding :-/
--
Steve Nygard
On Feb 8, 2007, at 7:37 AM, Gregory John Casamento wrote:
All,
My apologies to the group for not being able to participate more.
GNUspeech is a fascinating project. Since I took over as Chief
Maintainer for GNUstep, I've been extremely busy both with things
concerning GNUstep and other things for my work.
I am happy to answer any questions about work I did for GNUspeech,
and I promise, but can't guarantee that I will spend some time on it.
One of the things I felt is most important is to refactor it so
that the tube model resides in a separate library. This way, it
could be used by any number of things. Also, I'm wondering if it
would be possible to abstract the output so that the sounds
generated could be piped to a number of different outputs instead
of depending on the CoreAudio framework. This would mean that you
would, essentially, have an audio "Backend" for GNUspeech.
Anyway, I have been thinking about this project (as you can tell
from the above)... if only there were 36 hours in the day, instead
of 24.
Later, GJC
--
Gregory Casamento
_______________________________________________
gnuspeech-contact mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/gnuspeech-contact
