Hi Ken,
Monet has been ported to Mac OS/X and can be used to play with speech
postures, convolution rules, and explicit adjustment of intonation
patterns. The GNUStep port is under way, and held up because the Mac
port uses Core Audio which is not (yet) part of GNUStep.
You could do some experimentation to explore possibilities and learn
the core of the system by running Monet on Mac OS/X.
The "Synthesizer" port to Mac OS/X is well under way and will be
available "real soon now". Best guess is of the order of 6 weeks.
The GNUStep port of "Synthesizer" will have the same problem as Monet
-- namely dealing with the sound output properly within the GNUStep
framework.
Greg Casamento (GNUStep guru) is looking at porting the needed parts
of Core Audio to GNUStep, which would solve the above problems.
"Real-time Monet" -- an engine that will run **without interaction**
using the rules, postures, dictionaries and so on established by use
of Monet, Synthesizer, and associated databases to provide a speech
service for whatever language was embodied in the databases, in
whatever application or operating system facility wished to use
speech based on a standard API, will be ported to Mac OS/X as soon as
there is a stable version of Synthesizer. No sure guesses yet as to
how long that would take but, with luck and hard work probably before
the end of the year.
If you want immediate access to the whole set or tools, applications
dictionaries etc you can obtain black NeXT hardware from http://
www.blackholeinc.com and load up the software available from the
savannah gnuspeech site. I can give some information about
configuration, tools needed, where the passwords are for access to
the kits, tools that are not on the site that I could provide (very
few, mainly to do with NeXT dictionary creation). I should probably
put a page up about this, but not many people are wanting that info
and I really want to focus on Mac OS/X, GNU/Linux and GNUStep. Going
back to the NeXT implementation is deprecated, but is a possibility
for those in urgent need.
Hope that helps.
david
----
On Mar 5, 2006, at 10:20 AM, Ken Beesley wrote:
I haven't looked into gnuspeech for quite a while.
Could some kind soul post a brief update on the status
of the port? Has the "Synthesizer" (the GUI front-end
to tube) been ported?
My need is to produce a preliminary TTS system
for an American Indian language, without using any
recordings. The orthography is a reliable representation
of the phonology, so there is no need (at least initially) for
any kind of pronunciation dictionary, and I have already
done grapheme-to-phoneme conversion (IPA output).
My preferred platform is Apple OS X.
Thanks,
Ken
_______________________________________________
gnuspeech-contact mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/gnuspeech-contact
_______________________________________________
gnuspeech-contact mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/gnuspeech-contact
