[gnuspeech-contact] Re: Membership request for group Articulatory Speech Synthesis

[David Hill]

Hi Rafal,

As administrator of the gnuspeech project, I received your request  
for membership of the group.  You may not realise that it is not  
necessary to be a group member in order to download everything in the  
project CVS repository.  Group membership is intended for those who  
are actively involved in developing/maintaining the code and  
developing new modules, so that they can update the repository.

For obvious reasons, we cannot simply allow new people we do not know  
to have the ability to change what is in the repository since we have  
no knowledge of their skills, abilities, goals, and experience.  If  
you do work on existing code and make improvements/additions, or if  
you come up with new modules related to what exists, or to the aims  
of the project then, after an assessment of your work, we would be in  
a position to admit you as a group member with repository change  
privileges.

I greatly appreciate your interest in the work, and we are always  
short of good people to contribute to the considerable task of  
further development.

I don't know how much of the web site you have explored but you  
should at least check out:

http://www.gnu.org/software/gnuspeech/

(which is linked from the main page at: http://savannah.gnu.org/ 
projects/gnuspeech as the first item under "Quick Overview.)

It explains the structure and status of the project.  Under the third  
heading "Development Tools" you will find "- Browse Sources  
Repository" with which you can find current Macintosh Objective-C  
sources, as well as archival material from the NeXT implementation  
(the "TextToSpeech Kit" which was a fully functional commercially  
available real-time articulatory speech synthesis package for the  
NeXT.  With the demise of NeXT and its takeover by Apple, we put all  
the software and development tools into the gnuspeech project under a  
GPL so that it may be used freely under the terms of that licence by  
anyone).  We are currently developing a version that will compile and  
run under GNU/Linux GnuStep and on the Macintosh under OS X, using  
conditional compilation.  So far the main language development tool  
"Monet" has been ported to the Macintosh and the port of  
"Synthesizer" -- which provides an interactive means of using the  
basic tube model that emulates the vocal tract for study and database  
development is 80% complete.  "Monet" does allow arbitrary text to be  
spoken, using the inappropriately-named "Gnuspeech" pre-processor to  
convert the text to the input format required by "Monet".

After "Synthesiser" is completed, the two most important tools for  
new language development ("Monet" and "Synthesizer" will be available  
at least for the Macintosh under OS X.  The GnuStep version is more  
of a problem as the audio output facilities for GnuStep are non- 
existent, other than using basic IOCTL.  There is no equivalent of  
Core Audio, though Greg Casamento -- the GnuStep project leader has  
promised that such facilities will be added, but is too busy to  
undertake the work himself.

There is more information in the first URL I gave you above, and I  
would be very happy to provide help if you need it.  Please keep in  
touch and let me know how your own research is going.  Perhaps in the  
near future, you will become more closely involved in the gnuspeech  
work, and hopefully you will be able to join us as a full project  
member, once we have got to know you and your work.

Do you have access to a Macintosh running OS X 10.4 or 10.5?  You can  
download Monet, and the pre-processor Gnuspeech from the repository ,  
compile it and run it (there will be some small changes needed to  
deal with the ongoing updates to 10.4 and 10.5).  I could also send  
you a .tar file of the current 0.8 version of "Synthesizer" so you  
can get a feel for it.  It compiles and runs under 10.4.11/Xcode  
2.4.1/SDK 4 all the way to 10.5.4/Xcode 3.1/SDK5).  Alternatively, if  
you have access to a NeXT computer, you can get the entire developer  
TextToSpeech kit and simply run it.  There's a set of the required  
passwords under the "private" directory in the repository.

I look forward to hearing from you and wish you well with your PhD  
thesis.

All good wishes.

david

--------

On Sep 5, 2008, at 10:30 AM, Rafal wrote:

>         
> Rafal requested membership to the group Articulatory Speech Synthesis
> at Savannah
>
> User Details:
> -------------
> Name:   Rafal
> Login:   rpietruc
> Email:   address@hidden
>
> Project Details:
> ----------------
> Name:   Articulatory Speech Synthesis
> Unix Name: gnuspeech
> Main Page:        https://savannah.gnu.org/projects/gnuspeech
>
> Message from user:
> ------------------
>
> I'm preparing PhD thesis on Warsaw Univ of Technology on speech  
> analysis, I'm
> interested in open S/W for background work and I'm searching for  
> useful tools
> to not implement everthing by myself.
>
>          Rafal Pietruch
>
>
> Note:
> -----
> You receive this email because you are registered as an  
> administrator of this
> project and the system has been configured to send emails to  
> administrators
> when new users register.
>
>
>
>
> Please login and go to the page
> https://savannah.gnu.org/project/admin/useradmin.php?group=gnuspeech
> and approve or discard this membership request.
>
>
>  -- the Savannah team
>         
>
> _______________________________________________
>   Message sent via/by Savannah
>   http://savannah.gnu.org/