I was just wondering how to run lm3g2dmp so that it will get my language_model.arpaformat.DMP file in the right format and correcting this error message when I try to run ant hub4_trigram:
[java] Exception in thread "main" java.lang.Error: Bad binary LM file magic number: 67324752, not an LM dumpfile? [java] at edu.cmu.sphinx.linguist.language.ngram.large.BinaryLoader.readHeader(BinaryLoader.java:333) [java] at edu.cmu.sphinx.linguist.language.ngram.large.BinaryLoader.loadBinary(BinaryLoader.java:264) [java] at edu.cmu.sphinx.linguist.language.ngram.large.BinaryLoader.<init>(BinaryLoader.java:93) [java] at edu.cmu.sphinx.linguist.language.ngram.large.LargeTrigramModel.allocate(LargeTrigramModel.java:300) [java] at edu.cmu.sphinx.linguist.lextree.LexTreeLinguist.allocate(LexTreeLinguist.java:405) [java] at edu.cmu.sphinx.decoder.search.WordPruningBreadthFirstSearchManager.allocate(WordPruningBreadthFirstSearchManager.java:323
) [java] at edu.cmu.sphinx.decoder.Decoder.allocate(Decoder.java:109) [java] at edu.cmu.sphinx.recognizer.Recognizer.allocate(Recognizer.java:182) [java] at edu.cmu.sphinx.tools.batch.BatchModeRecognizer.decode(BatchModeRecognizer.java:244) [java] at edu.cmu.sphinx.tools.batch.BatchModeRecognizer.main(BatchModeRecognizer.java:628) [java] Java Result: 1
The error message I get when I run lm3g2dmp is:
$lm3g2dmp language_model.arpaformat.DMP new
INFO: lm3g2dmp.c(708): Reading LM file language_model.arpaformat.DMP (name "")
FATAL_ERROR: "lm3g2dmp.c", line 385: No \data\ mark in LM file
Thanks,
-Danny
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I figured out the problem. The location property of the trigramModel (in hub4.config.xml) component was set to my HUB4_8gau_13dCep_16k_40mel_133Hz_6855Hz.jar file rather than the language_model.arpaformat.DMP file.
-Danny
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I was just wondering how to run lm3g2dmp so that it will get my language_model.arpaformat.DMP file in the right format and correcting this error message when I try to run ant hub4_trigram:
[java] Exception in thread "main" java.lang.Error: Bad binary LM file magic number: 67324752, not an LM dumpfile?
[java] at edu.cmu.sphinx.linguist.language.ngram.large.BinaryLoader.readHeader(BinaryLoader.java:333)
[java] at edu.cmu.sphinx.linguist.language.ngram.large.BinaryLoader.loadBinary(BinaryLoader.java:264)
[java] at edu.cmu.sphinx.linguist.language.ngram.large.BinaryLoader.<init>(BinaryLoader.java:93)
[java] at edu.cmu.sphinx.linguist.language.ngram.large.LargeTrigramModel.allocate(LargeTrigramModel.java:300)
[java] at edu.cmu.sphinx.linguist.lextree.LexTreeLinguist.allocate(LexTreeLinguist.java:405)
[java] at edu.cmu.sphinx.decoder.search.WordPruningBreadthFirstSearchManager.allocate(WordPruningBreadthFirstSearchManager.java:323
)
[java] at edu.cmu.sphinx.decoder.Decoder.allocate(Decoder.java:109)
[java] at edu.cmu.sphinx.recognizer.Recognizer.allocate(Recognizer.java:182)
[java] at edu.cmu.sphinx.tools.batch.BatchModeRecognizer.decode(BatchModeRecognizer.java:244)
[java] at edu.cmu.sphinx.tools.batch.BatchModeRecognizer.main(BatchModeRecognizer.java:628)
[java] Java Result: 1
The error message I get when I run lm3g2dmp is:
$lm3g2dmp language_model.arpaformat.DMP new
INFO: lm3g2dmp.c(708): Reading LM file language_model.arpaformat.DMP (name "")
FATAL_ERROR: "lm3g2dmp.c", line 385: No \data\ mark in LM file
Thanks,
-Danny
I figured out the problem. The location property of the trigramModel (in hub4.config.xml) component was set to my HUB4_8gau_13dCep_16k_40mel_133Hz_6855Hz.jar file rather than the language_model.arpaformat.DMP file.
-Danny
There is no need to convert a dmp file to a dmp file. (that's what lm3g2dmp for.)
-gj
Thanks gj, you confirmed my suspicions. Anyone have ideas on how to resolve the error message regarding the bad binary LM file?