org.ow2.weblab.service.transcript.sphinx
Class WebLabTextTranscriptCreator

java.lang.Object
  org.ow2.weblab.service.transcript.sphinx.WebLabTextTranscriptCreator

public class WebLabTextTranscriptCreator
extends java.lang.Object

Field Summary

protected org.ow2.weblab.content.api.ContentManager	contentManager
protected SphinxTranscriptor	sphinxTranscriptor
protected java.lang.String	targetLang
protected boolean	writeFiller
protected boolean	writePronunciation
protected boolean	writeScore
protected boolean	writeTokenInfo

Constructor Summary

WebLabTextTranscriptCreator(SphinxTranscriptor sphinxTranscriptor, org.ow2.weblab.content.api.ContentManager contentManager, boolean writeTokenInfo, boolean writeFiller, boolean writeScore, boolean writePronunciation, java.lang.String targetLang)
Generates all text transcript from all audio media unit contained in a WebLab document using a specific sphinxTranscriptor

Method Summary

protected void	addWord(java.lang.StringBuffer sb, org.ow2.weblab.core.model.Annotation textAnnotation, org.ow2.weblab.core.model.Annotation audioAnnotation, org.ow2.weblab.core.model.Text text, org.ow2.weblab.core.model.Audio audio, edu.cmu.sphinx.decoder.search.Token token, edu.cmu.sphinx.frontend.FloatData startFeature, edu.cmu.sphinx.frontend.FloatData endFeature) Generates the two segments with all configured annotations
protected void	generatedTranscriptedText(edu.cmu.sphinx.decoder.search.Token curToken, org.ow2.weblab.core.model.Text text, org.ow2.weblab.core.model.Audio audio, org.ow2.weblab.core.model.Annotation textAnnotation, org.ow2.weblab.core.model.Annotation audioAnnotation) Creates aligned segments from Token, Text and Audio.
java.lang.String	toString()
void	transcriptDocument(org.ow2.weblab.core.model.Document parent)

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

Field Detail

sphinxTranscriptor

protected SphinxTranscriptor sphinxTranscriptor

writeTokenInfo

protected boolean writeTokenInfo

writeFiller

protected boolean writeFiller

writeScore

protected boolean writeScore

writePronunciation

protected boolean writePronunciation

targetLang

protected java.lang.String targetLang

contentManager

protected org.ow2.weblab.content.api.ContentManager contentManager

Constructor Detail

WebLabTextTranscriptCreator

public WebLabTextTranscriptCreator(SphinxTranscriptor sphinxTranscriptor,
                                   org.ow2.weblab.content.api.ContentManager contentManager,
                                   boolean writeTokenInfo,
                                   boolean writeFiller,
                                   boolean writeScore,
                                   boolean writePronunciation,
                                   java.lang.String targetLang)

Generates all text transcript from all audio media unit contained in a WebLab document using a specific sphinxTranscriptor

Parameters:

sphinxTranscriptor - model used to transcript

contentManager - content manager to be used

writeTokenInfo - true if your want to add extra information on each token

writeFiller - true if you want filler words

writeScore - true if you want to write score

writePronunciation - true if you want to write pronunciation

targetLang - target language

Method Detail

transcriptDocument

public void transcriptDocument(org.ow2.weblab.core.model.Document parent)
                        throws org.ow2.weblab.core.services.InvalidParameterException,
                               org.ow2.weblab.core.services.ContentNotAvailableException

Throws:

org.ow2.weblab.core.services.InvalidParameterException

org.ow2.weblab.core.services.ContentNotAvailableException

generatedTranscriptedText

protected void generatedTranscriptedText(edu.cmu.sphinx.decoder.search.Token curToken,
                                         org.ow2.weblab.core.model.Text text,
                                         org.ow2.weblab.core.model.Audio audio,
                                         org.ow2.weblab.core.model.Annotation textAnnotation,
                                         org.ow2.weblab.core.model.Annotation audioAnnotation)

Creates aligned segments from Token, Text and Audio. inspired from Sphinx code in Result.getTimedWordPath

Parameters:

token - Sphinx Token

text - Text media unit transcripted

audio - Audio source media unit

audioAnnotation - Annotation on Audio unit

textAnnotation - Annotation on Text unit

addWord

protected void addWord(java.lang.StringBuffer sb,
                       org.ow2.weblab.core.model.Annotation textAnnotation,
                       org.ow2.weblab.core.model.Annotation audioAnnotation,
                       org.ow2.weblab.core.model.Text text,
                       org.ow2.weblab.core.model.Audio audio,
                       edu.cmu.sphinx.decoder.search.Token token,
                       edu.cmu.sphinx.frontend.FloatData startFeature,
                       edu.cmu.sphinx.frontend.FloatData endFeature)

Generates the two segments with all configured annotations

Parameters:

textAnnotation - Annotation where to add text segment metadata

audioAnnotation - Annotation where to add audio segment metadata

text - Text unit to add linear segment to

audio - Audio unit to add temporal segment to

token - the Sphinx recognnized token

startFeature - Sphinx started feature

endFeature - Sphinx ending feature

toString

public java.lang.String toString()

Overrides:

toString in class java.lang.Object