I just realized I posted this in the wrong forum, the open discussion rather
than help forum; apologies for the cross posting:
I am using the transcriber demo with the WSJ 8k acoustic model and a custom
language model. when I run the jar I get:
error! java.lang.NullPointerException java.lang.NullPointerException at edu.cm
u.sphinx.model.acoustic.WSJ8gau13dCep8kHz31mel200Hz3500Hz.Model.hasNonSilenceF
iller(Model.java:565) at edu.cmu.sphinx.model.acoustic.WSJ8gau13dCep8kHz31mel2
00Hz3500Hz.Model.getHMMInSilenceContext(Model.java:518) at edu.cmu.sphinx.mode
l.acoustic.WSJ8gau13dCep8kHz31mel200Hz3500Hz.Model.lookupNearestHMM(Model.java
:229) at edu.cmu.sphinx.linguist.util.HMMPool.(HMMPool.java:101) at edu.cmu.sp
hinx.linguist.lextree.LexTreeLinguist.generateHmmTree(LexTreeLinguist.java:364
) at edu.cmu.sphinx.linguist.lextree.LexTreeLinguist.compileGrammar(LexTreeLin
guist.java:353) at edu.cmu.sphinx.linguist.lextree.LexTreeLinguist.allocate(Le
xTreeLinguist.java:279) at edu.cmu.sphinx.decoder.search.WordPruningBreadthFir
stSearchManager.allocate(WordPruningBreadthFirstSearchManager.java:211) at
edu.cmu.sphinx.decoder.AbstractDecoder.allocate(AbstractDecoder.java:67) at
edu.cmu.sphinx.recognizer.Recognizer.allocate(Recognizer.java:157) at
edu.cmu.sphinx.demo.transcriber.Transcriber.main(Transcriber.java:65)
The funny thing is I swear it used to work before... I ran it in september for
a demo and it worked fine.
I am using sphinx beta 2. It doesn't have this exception unless I use the WSJ
8k model loader. Seems to work with hub4 model and modelloader just fine. Also
the wsj 16k seems not to have this exception although when I sub in the wsj
16k version in the config file (and use 16k audio input files) it doesn't seem
to recognize anything, but it doesn't crash:)
The config file can be viewed from my accidental post in the open forum :
[https://sourceforge.net/projects/cmusphinx/forums/forum/382337/topic/1164455]
(https://sourceforge.net/projects/cmusphinx/forums/forum/382337/topic/1164455)
could be relevant.
Also upgrade to beta3.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
What would be the easiest way to upgrade my existing system to beta3? I have
modified lots of the examples, made my own models, etc. and don't want to lose
my work.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I just realized I posted this in the wrong forum, the open discussion rather
than help forum; apologies for the cross posting:
I am using the transcriber demo with the WSJ 8k acoustic model and a custom
language model. when I run the jar I get:
error! java.lang.NullPointerException java.lang.NullPointerException at edu.cm
u.sphinx.model.acoustic.WSJ8gau13dCep8kHz31mel200Hz3500Hz.Model.hasNonSilenceF
iller(Model.java:565) at edu.cmu.sphinx.model.acoustic.WSJ8gau13dCep8kHz31mel2
00Hz3500Hz.Model.getHMMInSilenceContext(Model.java:518) at edu.cmu.sphinx.mode
l.acoustic.WSJ8gau13dCep8kHz31mel200Hz3500Hz.Model.lookupNearestHMM(Model.java
:229) at edu.cmu.sphinx.linguist.util.HMMPool.(HMMPool.java:101) at edu.cmu.sp
hinx.linguist.lextree.LexTreeLinguist.generateHmmTree(LexTreeLinguist.java:364
) at edu.cmu.sphinx.linguist.lextree.LexTreeLinguist.compileGrammar(LexTreeLin
guist.java:353) at edu.cmu.sphinx.linguist.lextree.LexTreeLinguist.allocate(Le
xTreeLinguist.java:279) at edu.cmu.sphinx.decoder.search.WordPruningBreadthFir
stSearchManager.allocate(WordPruningBreadthFirstSearchManager.java:211) at
edu.cmu.sphinx.decoder.AbstractDecoder.allocate(AbstractDecoder.java:67) at
edu.cmu.sphinx.recognizer.Recognizer.allocate(Recognizer.java:157) at
edu.cmu.sphinx.demo.transcriber.Transcriber.main(Transcriber.java:65)
The funny thing is I swear it used to work before... I ran it in september for
a demo and it worked fine.
I am using sphinx beta 2. It doesn't have this exception unless I use the WSJ
8k model loader. Seems to work with hub4 model and modelloader just fine. Also
the wsj 16k seems not to have this exception although when I sub in the wsj
16k version in the config file (and use 16k audio input files) it doesn't seem
to recognize anything, but it doesn't crash:)
The config file can be viewed from my accidental post in the open forum :
[https://sourceforge.net/projects/cmusphinx/forums/forum/5470/topic/3449870]
(https://sourceforge.net/projects/cmusphinx/forums/forum/5470/topic/3449870)
[https://sourceforge.net/projects/cmusphinx/forums/forum/382337/topic/1164455]
(https://sourceforge.net/projects/cmusphinx/forums/forum/382337/topic/1164455)
could be relevant.
Also upgrade to beta3.
What would be the easiest way to upgrade my existing system to beta3? I have
modified lots of the examples, made my own models, etc. and don't want to lose
my work.
It should not be a problem.Just use the new sphinx4 jar. So download it,
change the classpaths to use it so when you run your app it uses the new one.