CMU Sphinx兼容词典

时间:2016-08-17 02:35:04

标签: speech-to-text cmusphinx

当针对提供的wav文件运行CMU Sphinx时,我收到此错误:

SEVERE: Missing HMM for unit P with lc=R rc=ER0
19:06:29.696 SEVERE lexTreeLinguist    Bad HMM Unit: EH1
Aug 16, 2016 7:06:29 PM edu.cmu.sphinx.linguist.lextree.HMMTree addPronunciation
SEVERE: Missing HMM for unit N with lc=EH1 rc=Z
19:06:29.697 SEVERE lexTreeLinguist    Bad HMM Unit: OW0
Exception in thread "main" java.lang.NullPointerException
at edu.cmu.sphinx.linguist.lextree.HMMNode.getBaseUnit(HMMTree.java:494)
at edu.cmu.sphinx.linguist.lextree.HMMNode.<init>(HMMTree.java:472)
at edu.cmu.sphinx.linguist.lextree.Node.addSuccessor(HMMTree.java:164)
at edu.cmu.sphinx.linguist.lextree.HMMTree$EntryPoint.createEntryPointMap(HMMTree.java:1154)
at edu.cmu.sphinx.linguist.lextree.HMMTree$EntryPointTable.createEntryPointMaps(HMMTree.java:1012)
at edu.cmu.sphinx.linguist.lextree.HMMTree.compile(HMMTree.java:784)
at edu.cmu.sphinx.linguist.lextree.HMMTree.<init>(HMMTree.java:706)
at edu.cmu.sphinx.linguist.lextree.LexTreeLinguist.generateHmmTree(LexTreeLinguist.java:428)
at edu.cmu.sphinx.linguist.lextree.LexTreeLinguist.compileGrammar(LexTreeLinguist.java:416)
at edu.cmu.sphinx.linguist.lextree.LexTreeLinguist.allocate(LexTreeLinguist.java:335)
at edu.cmu.sphinx.decoder.search.WordPruningBreadthFirstSearchManager.allocate(WordPruningBreadthFirstSearchManager.java:243)
at edu.cmu.sphinx.decoder.AbstractDecoder.allocate(AbstractDecoder.java:103)
at edu.cmu.sphinx.recognizer.Recognizer.allocate(Recognizer.java:164)
at edu.cmu.sphinx.api.StreamSpeechRecognizer.startRecognition(StreamSpeechRecognizer.java:52)
at edu.cmu.sphinx.api.StreamSpeechRecognizer.startRecognition(StreamSpeechRecognizer.java:39)

我正在使用the CMUDict on Github和语言模型on sourceforge

当googlin&#39;这个错误,人们暗示声学模型和字典之间存在不匹配。但是,我无法找到有关哪些型号/词典兼容的任何文档。 CMU网站不提供任何指导。我已尝试过多次配对,但我很感激直接指导。

1 个答案:

答案 0 :(得分:0)

与模型兼容的词典已包含在发行版和source tree中。

github项目中的字典在音素中有压力标记,在使用解码器之前必须先用脚本删除。