For the XO, it would be great if dictionaries were *not* constant databases. Many of the countries where the XO is/will be deployed have local languages for which good free electronic dictionaries very much do not exist. Moreover, even in places where the local language is digital-age, creating a schoolwide jargon file would be a great project for involving the kids.
I know that, if the University of Southern Denmark and the dead author's daughter ever finish giving me CC-like rights to a multilanguage Mayan dictionary, a good wiktionary-like activity would become *the* major argument for adopting the XO in Guatemala. (the dictionary is large, but antiquated - it would need to be user-extensible to be useful). I am very happy that a dictionary activity exists, and I know that the programming work for a user-modifiable one is several times as much as just a static one. So I am not belittling Chris's accomplishment, just saying that tying it to a static database is not the direction I would advise. Jameson On Mon, Mar 10, 2008 at 3:01 AM, Assim Deodia <[EMAIL PROTECTED]> wrote: > Hi, > > On 3/10/08, Hemant Goyal <[EMAIL PROTECTED]> wrote: > > > > Hi, > > > > > > > Next steps: > > > > > > * I have a "Speak" button that I'd like to hook up to espeak with > > > the > > > appropriate accent loaded. Josh, any interest in helping to get > > > that working? Should I just wait for the speech server? > > > > > > The speech dispatcher API is what you can use for the speech synthesis. > > The python API is accessible here : > > http://cvs.freebsoft.org/repository/speechd/src/python/speechd/client.py?view=markup > > > > I am still trying to get the speech dispatcher and its dependencies > > packaged for Fedora :\. And that is the only reason why the speech server is > > still not present on the XO :(. I really request an expert to look into this > > and help speed up the work. > > > > In the long run, we plan to explore other Text to Speech engines other > > than eSpeak (because of voice quality issues). So coupling your app with > > eSpeak is not so sensible when you can directly use speech-dispatcher for > > all testing. > > > > I have a few unapproved speech-dispatcher packages here: > > http://www.nsitonline.in/hemant/stuff/speechd-rpm/ > > > > You'll need dotconf packages to install speech dispatcher. Perhaps Assim > > can help you get a *usable* dotconf pacakge. > > > > You can get dotconf RPM's from > > > http://nsitonline.in/assim/stuffs/olpc/dotconf/dotconf-devel-1.0.13-6.fc7.i386.rpm > > http://nsitonline.in/assim/stuffs/olpc/dotconf/dotconf-debuginfo-1.0.13-6.fc7.i386.rpm > > http://nsitonline.in/assim/stuffs/olpc/dotconf/dotconf-1.0.13-6.fc7.i386.rpm > > These RPM's are clean to use with only some minor bugs regarding Fedora > Guidelines. We would really appreciate some help in process of review > request to get both the packages > (speech-dispatcher and dotconf) into OLPC build soon. > > > I'd gladly hook your activity to speech dispatcher and make it self > > voicing _once_ speechd gets accepted in the OLPC build. > > > > Thanks! > > Hemant > > > > > -- > Regards > Assim > _______________________________________________ > Sugar mailing list > [email protected] > http://lists.laptop.org/listinfo/sugar > >
_______________________________________________ Sugar mailing list [email protected] http://lists.laptop.org/listinfo/sugar
