Helmut,<br><br>Your hard work is appreciated. Like Brian said, we'd all be interested in knowing more. Please feel free to put this on the wiki or see me off list and we'll discuss further how to document it for the good of the FS community.<br>
<br>Thanks!<br>-MC<br><br><div class="gmail_quote">On Fri, Jul 31, 2009 at 6:18 AM, Helmut Kuper <span dir="ltr"><<a href="mailto:helmut.kuper@ewetel.de">helmut.kuper@ewetel.de</a>></span> wrote:<br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
-----BEGIN PGP SIGNED MESSAGE-----<br>
Hash: SHA1<br>
<br>
Hello,<br>
<br>
I spent a few days working on my problem. I played around with<br>
voxforge's data, read some of the train docs of CMU, and debugged the<br>
sphinxbase, pocketsphinx and mod_pocketsphinx.<br>
<br>
Results:<br>
- -I found a way to use the voxforge data as training data for creation of<br>
a german language corpus.<br>
<br>
- -I enabled the logging of pocketsphinx to stderr (Dirty, but easy way to<br>
see what went wrong when FS loads grammar, mdef, etc and simply stopped.<br>
Very helpful!<br>
<br>
- -I had to add a "dictcase" parameter to pocketsphinx.conf.xml resp<br>
mod_pocketsphinx.c to allow case sensitive dictionaries (like the german<br>
dictionary from voxforge).<br>
<br>
<br>
FS starts up with german language model and detects the words as<br>
expected. But it's not so reliable as I want to ... I guess this is<br>
caused by the very small amount of training audio data. I used 4000 of<br>
19000 audio files provided by voxforge due to the reason that voyforge's<br>
training fileid-list contains only 4000 files ... I have to create new<br>
fileid-list and transcription-lists containing all audio I have<br>
downloaded from voxforge.<br>
<br>
<br>
Quite complex the whole thing ...<br>
<br>
regards<br>
helmut<br>
<br>
<br>
<br>
On 10.07.2009 14:53, Helmut Kuper wrote:<br>
> Hi,<br>
><br>
> I try to change pocketsphinx's grammar from default (english) to german.<br>
> I found this archive<br>
> (<a href="http://www.repository.voxforge1.org/downloads/de/Trunk/AcousticModels/" target="_blank">http://www.repository.voxforge1.org/downloads/de/Trunk/AcousticModels/</a>), which<br>
> contains similar files like those which can be found in<br>
> grammar/model/communicator directory.<br>
><br>
> Unfortunately FS crashed without writing a core file nor logfile enries<br>
> as soon as as pizza demo trys to detect speech.<br>
><br>
> Any Ideas? Maybe someone has already working grammar/model files for<br>
> german language?<br>
><br>
><br>
> regards<br>
> helmut<br>
><br>
-----BEGIN PGP SIGNATURE-----<br>
Version: GnuPG v1.4.7 (MingW32)<br>
<br>
iD8DBQFKcu8z4tZeNddg3dwRAiVKAKCJdAvZyYXLClWp+kQO4MhsPVzbSQCgoX/T<br>
7NosCujHwpkCXHzxkEY/N3M=<br>
=6Bu8<br>
-----END PGP SIGNATURE-----<br>
<br>
_______________________________________________<br>
FreeSWITCH-users mailing list<br>
<a href="mailto:FreeSWITCH-users@lists.freeswitch.org">FreeSWITCH-users@lists.freeswitch.org</a><br>
<a href="http://lists.freeswitch.org/mailman/listinfo/freeswitch-users" target="_blank">http://lists.freeswitch.org/mailman/listinfo/freeswitch-users</a><br>
UNSUBSCRIBE:<a href="http://lists.freeswitch.org/mailman/options/freeswitch-users" target="_blank">http://lists.freeswitch.org/mailman/options/freeswitch-users</a><br>
<a href="http://www.freeswitch.org" target="_blank">http://www.freeswitch.org</a><br>
</blockquote></div><br>