[mary-dev] adding new NLPs to the system

[fxavier at ircam.fr]

Yes I tried to write it into the marybase.config, but same problem.
Actually I didn't say that the conversion TEXT->RAWMARYXML failed, of
course it works with every kind of local because it only uses the general
class TextToMaryXML included in marybase.config.
But TEXT->PHONEMES doesnt work, and according to what's on the server.log,
It stops right after the TEXT->RAWMARYXML.

2011-06-21 11:49:29,827 [I/O dispatcher 5] INFO  marytts.R 2 -
>> TextToMaryXML (marytts.modules.TextToMaryXML)
>> 2011-06-21 11:49:29,827 [I/O dispatcher 5] INFO  marytts.R 2 Next
>> module:
>> TextToMaryXML
>> 2011-06-21 11:49:29,828 [I/O dispatcher 5] ERROR marytts.server
>> Processing
>> failed.

So I tried the conversion RAWMARYXML->PHONEMES, which is the input and
output of my class PhonemiserFR, so that it is the only class used in
theory, but I have diectly the ERROR marytts.server. Which means that the
class is not even recognized. As you can see in this portion, the class
PhonemiserFR is never mentionned. Even if my code is not good, at least it
should be written in the server.log, and after only, must say that
processing failed, doesn'it?
I think the problem comes from my fr.config which, without being ignored
(for example when including some dummy class or forgetting some arguments,
the server throws an exception), is not properly recognized.



Florent






> One thing you could try for testing is to write the
> 	marytts.language.fr.PhonemiserFR \
> directly into marybase.config -- just to make sure the problem is not
> with fr.config being ignored.
>
> How can the TEXT -> RAWMARYXML conversion fail? Even with a dummy locale
> "bla" it is supposed to work:
> http://mary.dfki.de:59125/process?INPUT_TEXT=Hello+world&INPUT_TYPE=TEXT&OUTPUT_TYPE=RAWMARYXML&LOCALE=bla
>
> M.
>
> On 21.06.11 12:10, fxavier at ircam.fr wrote:
>> Hello Nickolay,
>>
>> Thank you for your answer. I did install all the languages (not the
>> voices) and it works... it's a real stupid mistake I must admit.
>>
>> However, still I 'm having problem with french. I guess it is not a
>> problem if the voice is not installed since it is just a warning. For
>> the
>> fr.config file, now it just look like this:
>>
>>
>>
>> ##########################################################################
>> # MARY TTS configuration file fr.config
>> ##########################################################################
>>
>> name = fr
>> fr.version = 4.3.0
>>
>> provides = a-language
>>
>> requires = \
>>      marybase
>>
>>
>> ###########################################################################
>> ############################## The Modules
>> ###############################
>> ###########################################################################
>>
>> modules.classes.list = \
>> 	marytts.language.fr.PhonemiserFR \
>>
>>
>>
>>
>> This class PhonemiserFR takes as input a RAWMARYXML and output PHONEMES.
>> I
>> included some logger.debug() inside the constructor and the process
>> method, but nothing is written into the server.log file. What I deduce
>> from the exceptions is that the modules TextToMaryXML and PhonemiserFR
>> are
>> not properly interconnected, so I'm missing something, maybe a file to
>> tweak, or some includes missing? However, the output of TextToMaryXML is
>> a
>> RAWMARYXML, which is the input of the PhonemiserFR class. I think I've
>> changed the build.xml file so it is correct now. As I can see in the
>> server.log:
>>
>>
>> 2011-06-21 11:49:29,827 [I/O dispatcher 5] INFO  marytts.R 2 -
>> TextToMaryXML (marytts.modules.TextToMaryXML)
>> 2011-06-21 11:49:29,827 [I/O dispatcher 5] INFO  marytts.R 2 Next
>> module:
>> TextToMaryXML
>> 2011-06-21 11:49:29,828 [I/O dispatcher 5] ERROR marytts.server
>> Processing
>> failed.
>>
>>
>> TextToMaryXML doesn't process.
>>
>>
>>
>> Florent
>>
>>
>>>
>>> Hello Florent
>>>
>>> This log says that you have no default voice defined for locale fr,
>>> this
>>> might be the reason of failure. Maybe you don't have German voice
>>> installed too and that's why it fails for German. Once you will install
>>> one German voice it should start working with German.
>>>
>>> Also in your config it seems you miss
>>>
>>> locale = fr
>>>
>>> In any case it's very easy to debug the specific issues if you can just
>>> add debugging print calls in the server call and track on which stage
>>> it
>>> fail.
>>>
>>>
>>
>> _______________________________________________
>> Mary-dev mailing list
>> Mary-dev at dfki.de
>> http://www.dfki.de/mailman/cgi-bin/listinfo/mary-dev
>
> --
> Dr. Marc Schröder, Senior Researcher at DFKI GmbH
> Project leader for DFKI in SSPNet http://sspnet.eu
> Team Leader DFKI TTS Group http://mary.dfki.de
> Editor W3C EmotionML Working Draft http://www.w3.org/TR/emotionml/
> Portal Editor http://emotion-research.net
>
> Homepage: http://www.dfki.de/~schroed
> Email: marc.schroeder at dfki.de
> Phone: +49-681-85775-5303
> Postal address: DFKI GmbH, Campus D3_2, Stuhlsatzenhausweg 3, D-66123
> Saarbrücken, Germany
> --
> Official DFKI coordinates:
> Deutsches Forschungszentrum fuer Kuenstliche Intelligenz GmbH
> Trippstadter Strasse 122, D-67663 Kaiserslautern, Germany
> Geschaeftsfuehrung:
> Prof. Dr. Dr. h.c. mult. Wolfgang Wahlster (Vorsitzender)
> Dr. Walter Olthoff
> Vorsitzender des Aufsichtsrats: Prof. Dr. h.c. Hans A. Aukes
> Amtsgericht Kaiserslautern, HRB 2313
> _______________________________________________
> Mary-dev mailing list
> Mary-dev at dfki.de
> http://www.dfki.de/mailman/cgi-bin/listinfo/mary-dev
>