[mary-dev] [mary-div] Adding new language support - Little help in running the feature marker program
Ingmar Steiner
ingmar.steiner at dfki.de
Mon Jan 5 10:44:44 CET 2015
Dear Tharaka,
the config file is generated by the `wkdb_setup.sh` script, but it
should be easy enough to look at that file and see the config file
format without actually running it:
https://raw.githubusercontent.com/marytts/marytts/bf487faa26619e0c26836a62433f2e767d7faa26/marytts-assembly/assembly-builder/src/builder/bin/wkdb_setup.sh
In case of problems, please open a new issue on GitHub.
Best wishes,
-Ingmar
On 02.01.15 12:02, Pubudu Tharaka wrote:
> Hi,
>
> I'm trying to add Sinhala language support to MaryTTS. Instead of
> extracting words from Wikipedia I took most frequent distinct words from
> a local text corpus. And created a mysql table as it is shown in the
> guide lines 2.2 having 3 columns id, word, frequency and put the words
> into the database. Then I ran the Transcription tool and created the
> letter to sound rule. Now I have successfully built and tested MaryTTS
> with Sinhala support. The next thing I have to do is Running feature
> maker with the minimal nlp components. According to the guide lines to
> do so I have to run the "wkdb_featuremaker.sh". To run that script it
> requests a config file as a command line argument. The config file is
> identified as "wkdb config file". As I skipped the Wikipedia process I
> think I have missed this config file. Is there any possibility that I
> can create this config file manually. If possible can someone please
> give me a sample file so that I can modify it to Sinhala language.
>
> Thank you.
>
> Regards,
> Tharaka.
>
>
> _______________________________________________
> Mary-dev mailing list
> Mary-dev at dfki.de
> http://www.dfki.de/mailman/cgi-bin/listinfo/mary-dev
>
More information about the Mary-dev
mailing list