Error running Pythontranscriber.py

Question

When I try to run pythontranscriber.py, it fails with the following error:

Traceback (most recent call last):
  File "PythonTranscriber.py", line 14, in <module>
  from edu.cmu.sphinx.decoder import Decoder
  ImportError: No module named edu

My script PythonTranscriber.py:

import sys

libDir = "/home/karen/sphinx4-1.0beta5-scr/sphinx4-1.0beta5/src/sphinx4/" 
classPaths = [
    "sphinx4.jar",
    "jsapi.jar" ] 
for classPath in classPaths:
    sys.path.append(libDir + classPath)

true = 1 
false = 0

from edu.cmu.sphinx.decoder import Decoder 
from edu.cmu.sphinx.decoder import ResultListener 
from edu.cmu.sphinx.decoder.pruner import SimplePruner 
from edu.cmu.sphinx.decoder.scorer import ThreadedAcousticScorer 
from edu.cmu.sphinx.decoder.search import PartitionActiveListFactory 
from edu.cmu.sphinx.decoder.search import SimpleBreadthFirstSearchManager 
from edu.cmu.sphinx.frontend import DataBlocker 
from edu.cmu.sphinx.frontend import FrontEnd 
from edu.cmu.sphinx.frontend.endpoint import NonSpeechDataFilter 
from edu.cmu.sphinx.frontend.endpoint import SpeechClassifier 
from edu.cmu.sphinx.frontend.endpoint import SpeechMarker 
from edu.cmu.sphinx.frontend.feature import DeltasFeatureExtractor 
from edu.cmu.sphinx.frontend.feature import LiveCMN 
from edu.cmu.sphinx.frontend.filter import Preemphasizer 
from edu.cmu.sphinx.frontend.frequencywarp import MelFrequencyFilterBank 
from edu.cmu.sphinx.frontend.transform import DiscreteCosineTransform 
from edu.cmu.sphinx.frontend.transform import DiscreteFourierTransform 
from edu.cmu.sphinx.frontend.util import AudioFileDataSource 
from edu.cmu.sphinx.frontend.window import RaisedCosineWindower 
from edu.cmu.sphinx.instrumentation import BestPathAccuracyTracker 
from edu.cmu.sphinx.instrumentation import MemoryTracker 
from edu.cmu.sphinx.instrumentation import SpeedTracker 
from edu.cmu.sphinx.jsapi import JSGFGrammar 
from edu.cmu.sphinx.linguist.acoustic import UnitManager 
from edu.cmu.sphinx.linguist.acoustic.tiedstate import Sphinx3Loader 
from edu.cmu.sphinx.linguist.acoustic.tiedstate import TiedStateAcousticModel 
from edu.cmu.sphinx.linguist.dictionary import FastDictionary 
from edu.cmu.sphinx.linguist.flat import FlatLinguist 
from edu.cmu.sphinx.recognizer import Recognizer 
from edu.cmu.sphinx.util import LogMath 
from java.util.logging import Logger 
from java.util.logging import Level 
from java.net import URL 
from java.util import ArrayList

# if (args.length < 1) {
#  throw new Error("USAGE: GroovyTranscriber <sphinx4 root> [<WAV file>]")
# }

root = "../../.."

# init common  
Logger.getLogger("").setLevel(Level.WARNING) 
logMath = LogMath(1.0001, true) 
absoluteBeamWidth = -1 
relativeBeamWidth = 1E-80 
wordInsertionProbability = 1E-36 
languageWeight = 8.0

# init audio data 
audioSource = AudioFileDataSource(3200, None) 
audioURL =  URL("file:" + root + "/src/apps/edu/cmu/sphinx/demo/transcriber/10001-90210-01803.wav")

# (args.length > 1) ?
#  File(args[0]).toURI().toURL() :   audioSource.setAudioFile(audioURL, None)

# init front end 
dataBlocker = DataBlocker(
        10 # blockSizeMs ) speechClassifier = SpeechClassifier(
        10,     # frameLengthMs,
        0.003, # adjustment,
        10,     # threshold,
        0       # minSignal 
)

speechMarker = SpeechMarker(
        200, # startSpeechTime,
        500, # endSilenceTime,
        100, # speechLeader,
        50,  # speechLeaderFrames
        100  # speechTrailer 
)

nonSpeechDataFilter = NonSpeechDataFilter()

premphasizer = Preemphasizer(
        0.97 # preemphasisFactor 
) 
windower = RaisedCosineWindower(
        0.46, # double alpha
        25.625, # windowSizeInMs
        10.0 # windowShiftInMs 
) 
fft = DiscreteFourierTransform(
        -1, # numberFftPoints
        false # invert 
) 
melFilterBank = MelFrequencyFilterBank(
        130.0, # minFreq,
        6800.0, # maxFreq,
        40 # numberFilters 
) 
dct = DiscreteCosineTransform(
        40, # numberMelFilters,
        13  # cepstrumSize 
) 
cmn = LiveCMN(
        12.0, # initialMean,
        100,  # cmnWindow,
        160   # cmnShiftWindow 
) 
featureExtraction = DeltasFeatureExtractor(
        3 # window 
)

pipeline = [
        audioSource,
        dataBlocker,
        speechClassifier,
        speechMarker,
        nonSpeechDataFilter,
        premphasizer,
        windower,
        fft,
        melFilterBank,
        dct,
        cmn,
        featureExtraction ]

frontend = FrontEnd(pipeline)

# init models 
unitManager = UnitManager()

modelLoader = Sphinx3Loader(
        "file:" + root + "/models/acoustic/tidigits/model.props",
        logMath,
        unitManager,
        true,
        true,
        39,
        "file:" + root + "/models/acoustic/tidigits/wd_dependent_phone.500.mdef",
        "file:" + root + "/models/acoustic/tidigits/wd_dependent_phone.cd_continuous_8gau/",
        0.0,
        1e-7,
        0.0001,
        true)

model = TiedStateAcousticModel(modelLoader, unitManager, true)

dictionary = FastDictionary(
        URL("file:" + root + "/models/acoustic/tidigits/dictionary"),
        URL("file:" + root + "/models/acoustic/tidigits/fillerdict"),
        ArrayList(),
        false,
        "<sil>",
        false,
        false,
        unitManager)

# init linguist 
grammar = JSGFGrammar(
        # URL baseURL,
        URL("file:" + root + "/src/apps/edu/cmu/sphinx/demo/transcriber/"),
        logMath, # LogMath logMath,
        "digits", # String grammarName,
        false, # boolean showGrammar,
        false, # boolean optimizeGrammar,
        false, # boolean addSilenceWords,
        false, # boolean addFillerWords,
        dictionary # Dictionary dictionary 
)

linguist = FlatLinguist(
        model, # AcousticModel acousticModel,
        logMath, # LogMath logMath,
        grammar, # Grammar grammar,
        unitManager, # UnitManager unitManager,
        wordInsertionProbability, # double wordInsertionProbability,
        1.0, # double silenceInsertionProbability,
        1.0, # double fillerInsertionProbability,
        1.0, # double unitInsertionProbability,
        languageWeight, # float languageWeight,
        false, # boolean dumpGStates,
        false, # boolean showCompilationProgress,
        false, # boolean spreadWordProbabilitiesAcrossPronunciations,
        false, # boolean addOutOfGrammarBranch,
        1.0, # double outOfGrammarBranchProbability,
        1.0, # double phoneInsertionProbability,
        None # AcousticModel phoneLoopAcousticModel 
)

# init recognizer 
scorer = ThreadedAcousticScorer(frontend, None, 10, true, 0)

pruner = SimplePruner()

activeListFactory = PartitionActiveListFactory(absoluteBeamWidth, relativeBeamWidth, logMath)

searchManager = SimpleBreadthFirstSearchManager(
        logMath, linguist, pruner,
        scorer, activeListFactory,
        false, 0.0, 0, false)

decoder = Decoder(searchManager,
        false, false,
        ArrayList(),
        100000)

recognizer = Recognizer(decoder, None)

# allocate the resourcs necessary for the recognizer recognizer.allocate()

# Loop unitl last utterance in the audio file has been decoded, in which case the recognizer will return None. 
result = recognizer.recognize() 
while (result != None):
    resultText = result.getBestResultNoFiller()
    print resultText
    result = recognizer.recognize()

I have Jython installed already!

Please help me!

(1) Are you actually running this in Jython? Just having it installed doesn't magically change CPython into Jython… (2) Do you have `edu.cmu.sphinx.decoder` installed? — abarnert, Nov 13 '12 at 02:13
How are you calling the script? You should be using `jython pythontranscriber.py`, or something to that effect. — David C, Nov 13 '12 at 02:34
no, am not doing this magically... im running this command: java -jar /home/karen/jython-installer-2.5.3/jython.jar PythonTranscriber.py — karensantana, Nov 13 '12 at 15:04
Have you tried adding sphinx4.jar and jsapi.jar to your classpath? `export CLASSPATH=/path/to/jar` — David C, Nov 13 '12 at 18:51
You can probably edit your post to just include all the lines through the last `from ... import x` statements. The rest isn't very relevant. — David C, Nov 15 '12 at 00:59

David C · Accepted Answer · 2012-11-15T06:37:33.290

1

To load Java code, try adding jars to your classpath.

If you're not familiar with how to do that, either use

 export CLASSPATH=$CLASSPATH:/path/to/sphinx4.jar:/path/jsapi.jar

Or add it to your command line script:

java -cp /path/to/jars -jar /home/karen/jython-installer-2.5.3/jython.jar PythonTranscriber.py

EDIT:

So, I followed the instructions here to download and install Sphinx-4. The path to the jar files was different than you had in your script. They were under the sphinx4/lib directory. You may want to double-check the correctness of your path.

I ran your script as below (modified, but no major changes):

import sys,os
lib_dir = '/home/.../jars/sphinx4/lib/'
classpaths = [ 'sphinx4.jar', 'jsapi.jar' ]
for cp in classpaths:
    sys.path.append(os.path.join(lib_dir,cp))

from edu.cmu.sphinx.decoder import Decoder
from ... import ...
etc.
from java.util import ArrayList

with the command:

jython python_transcriber.py

This almost worked, except that you have one more import error:
```
from edu.cmu.sphinx.jsapi import JSGFGrammar
```
should, according to the javadoc, be:
```
from edu.cmu.sphinx.jsgf import JSGFGrammar
```

That took care of it for me. Didn't try the rest of the code, but at least all the imports work.

edited Nov 15 '12 at 06:37

answered Nov 13 '12 at 19:01

David C

7,204
5
46
65

I did it but i got the same error. I think that jython does not find java's packages or sphinx4.jar. – karensantana Nov 14 '12 at 15:30
Are you sure you got the path correct? I installed Sphinx-4, and the jars defaulted to the `sphinx4/lib` directory. – David C Nov 15 '12 at 06:00
Also, make sure you have `sharutils` installed (required by Sphinx-4): `sudo apt-get install sharutils`. – David C Nov 15 '12 at 06:58
Thank you! All the imports worked. Am going to try the rest of the code. – karensantana Nov 16 '12 at 23:28
Look at this [http://stackoverflow.com/questions/13426763/error-executing-pythontranscriber-py] can you help me? – karensantana Nov 17 '12 at 22:59

Error running Pythontranscriber.py

1 Answers1