问题
Currently we are investigating into using sphinx4 for speech recognition. We are trying to achieve a good response for a dictation type application. The input is a wav file and we wish to transcribe it. I have looked into the LatticeDemo and Transcriber demo provided by Sphinx4. When i utilize the same configuration , the response is pretty poor. I have tried to tweak in the configuration files but it simply does not recognize the words. the transcriber demo provided is for digits, i have modified the config file to understand words. But i am not sure if i am missing something. I have attached the config file . Kindly suggest any improvements that can be made.
<config>
<!-- ******************************************************** -->
<!-- frequently tuned properties -->
<!-- ******************************************************** -->
<property name="absoluteBeamWidth" value="500"/>
<property name="relativeBeamWidth" value="1E-60"/>
<property name="absoluteWordBeamWidth" value="20"/>
<property name="relativeWordBeamWidth" value="1E-40"/>
<property name="wordInsertionProbability" value="1E-16"/>
<property name="languageWeight" value="7.0"/>
<property name="silenceInsertionProbability" value=".1"/>
<property name="frontend" value="epFrontEnd"/>
<property name="recognizer" value="recognizer"/>
<property name="showCreations" value="false"/>
<!-- ******************************************************** -->
<!-- word recognizer configuration -->
<!-- ******************************************************** -->
<component name="recognizer"
type="edu.cmu.sphinx.recognizer.Recognizer">
<property name="decoder" value="decoder"/>
<propertylist name="monitors">
<item>accuracyTracker </item>
<item>speedTracker </item>
<item>memoryTracker </item>
<item>recognizerMonitor </item>
</propertylist>
</component>
<!-- ******************************************************** -->
<!-- The Decoder configuration -->
<!-- ******************************************************** -->
<component name="decoder" type="edu.cmu.sphinx.decoder.Decoder">
<property name="searchManager" value="wordPruningSearchManager"/>
<property name="featureBlockSize" value="50"/>
</component>
<!-- ******************************************************** -->
<!-- The Search Manager -->
<!-- ******************************************************** -->
<component name="wordPruningSearchManager"
type="edu.cmu.sphinx.decoder.search.WordPruningBreadthFirstSearchManager">
<property name="logMath" value="logMath"/>
<property name="linguist" value="lexTreeLinguist"/>
<property name="pruner" value="trivialPruner"/>
<property name="scorer" value="threadedScorer"/>
<property name="activeListManager" value="activeListManager"/>
<property name="growSkipInterval" value="0"/>
<property name="checkStateOrder" value="false"/>
<property name="buildWordLattice" value="true"/>
<property name="acousticLookaheadFrames" value="1.7"/>
<property name="relativeBeamWidth" value="${relativeBeamWidth}"/>
</component>
<!-- ******************************************************** -->
<!-- The Active Lists -->
<!-- ******************************************************** -->
<component name="activeListManager"
type="edu.cmu.sphinx.decoder.search.SimpleActiveListManager">
<propertylist name="activeListFactories">
<item>standardActiveListFactory</item>
<item>wordActiveListFactory</item>
<item>wordActiveListFactory</item>
<item>standardActiveListFactory</item>
<item>standardActiveListFactory</item>
<item>standardActiveListFactory</item>
</propertylist>
</component>
<component name="standardActiveListFactory"
type="edu.cmu.sphinx.decoder.search.PartitionActiveListFactory">
<property name="logMath" value="logMath"/>
<property name="absoluteBeamWidth" value="${absoluteBeamWidth}"/>
<property name="relativeBeamWidth" value="${relativeBeamWidth}"/>
</component>
<component name="wordActiveListFactory"
type="edu.cmu.sphinx.decoder.search.PartitionActiveListFactory">
<property name="logMath" value="logMath"/>
<property name="absoluteBeamWidth" value="${absoluteWordBeamWidth}"/>
<property name="relativeBeamWidth" value="${relativeWordBeamWidth}"/>
</component>
<!-- ******************************************************** -->
<!-- The Pruner -->
<!-- ******************************************************** -->
<component name="trivialPruner"
type="edu.cmu.sphinx.decoder.pruner.SimplePruner"/>
<!-- ******************************************************** -->
<!-- TheScorer -->
<!-- ******************************************************** -->
<component name="threadedScorer"
type="edu.cmu.sphinx.decoder.scorer.ThreadedAcousticScorer">
<property name="frontend" value="${frontend}"/>
</component>
<!-- ******************************************************** -->
<!-- The linguist configuration -->
<!-- ******************************************************** -->
<component name="lexTreeLinguist"
type="edu.cmu.sphinx.linguist.lextree.LexTreeLinguist">
<property name="logMath" value="logMath"/>
<property name="acousticModel" value="wsj"/>
<property name="languageModel" value="trigramModel"/>
<property name="dictionary" value="dictionary"/>
<property name="addFillerWords" value="false"/>
<property name="fillerInsertionProbability" value="1E-10"/>
<property name="generateUnitStates" value="false"/>
<property name="wantUnigramSmear" value="true"/>
<property name="unigramSmearWeight" value="1"/>
<property name="wordInsertionProbability"
value="${wordInsertionProbability}"/>
<property name="silenceInsertionProbability"
value="${silenceInsertionProbability}"/>
<property name="languageWeight" value="${languageWeight}"/>
<property name="unitManager" value="unitManager"/>
</component>
<!-- ******************************************************** -->
<!-- The Dictionary configuration -->
<!-- ******************************************************** -->
<component name="dictionary"
type="edu.cmu.sphinx.linguist.dictionary.FastDictionary">
<property name="dictionaryPath"
value="resource:/WSJ_8gau_13dCep_16k_40mel_130Hz_6800Hz/dict/cmudict.0.6d"/>
<property name="fillerPath"
value="resource:/WSJ_8gau_13dCep_16k_40mel_130Hz_6800Hz/noisedict"/>
<property name="addSilEndingPronunciation" value="false"/>
<property name="wordReplacement" value="<sil>"/>
<property name="unitManager" value="unitManager"/>
</component>
<!-- ******************************************************** -->
<!-- The Language Model configuration -->
<!-- ******************************************************** -->
<component name="trigramModel"
type="edu.cmu.sphinx.linguist.language.ngram.large.LargeTrigramModel">
<property name="unigramWeight" value=".5"/>
<property name="maxDepth" value="3"/>
<property name="logMath" value="logMath"/>
<property name="dictionary" value="dictionary"/>
<property name="location"
value="./models/language/wsj/wsj5kc.Z.DMP"/>
</component>
<!-- ******************************************************** -->
<!-- The acoustic model configuration -->
<!-- ******************************************************** -->
<component name="wsj"
type="edu.cmu.sphinx.linguist.acoustic.tiedstate.TiedStateAcousticModel">
<property name="loader" value="wsjLoader"/>
<property name="unitManager" value="unitManager"/>
</component>
<component name="wsjLoader" type="edu.cmu.sphinx.linguist.acoustic.tiedstate.Sphinx3Loader">
<property name="logMath" value="logMath"/>
<property name="unitManager" value="unitManager"/>
<property name="location" value="resource:/WSJ_8gau_13dCep_16k_40mel_130Hz_6800Hz"/>
</component>
<!-- ******************************************************** -->
<!-- The unit manager configuration -->
<!-- ******************************************************** -->
<component name="unitManager"
type="edu.cmu.sphinx.linguist.acoustic.UnitManager"/>
<!-- ******************************************************** -->
<!-- The frontend configuration -->
<!-- ******************************************************** -->
<component name="epFrontEnd" type="edu.cmu.sphinx.frontend.FrontEnd">
<propertylist name="pipeline">
<item>audioFileDataSource </item>
<item>dataBlocker </item>
<item>speechClassifier </item>
<item>speechMarker </item>
<item>nonSpeechDataFilter </item>
<item>preemphasizer </item>
<item>windower </item>
<item>fft </item>
<item>melFilterBank </item>
<item>dct </item>
<item>liveCMN </item>
<item>featureExtraction </item>
</propertylist>
</component>
<component name="audioFileDataSource" type="edu.cmu.sphinx.frontend.util.AudioFileDataSource"/>
<component name="microphone"
type="edu.cmu.sphinx.frontend.util.Microphone">
<property name="closeBetweenUtterances" value="false"/>
</component>
<component name="dataBlocker" type="edu.cmu.sphinx.frontend.DataBlocker"/>
<component name="speechClassifier"
type="edu.cmu.sphinx.frontend.endpoint.SpeechClassifier">
<property name="threshold" value="13"/>
</component>
<component name="nonSpeechDataFilter"
type="edu.cmu.sphinx.frontend.endpoint.NonSpeechDataFilter"/>
<component name="speechMarker"
type="edu.cmu.sphinx.frontend.endpoint.SpeechMarker">
<property name="speechTrailer" value="50"/>
</component>
<component name="preemphasizer"
type="edu.cmu.sphinx.frontend.filter.Preemphasizer"/>
<component name="windower"
type="edu.cmu.sphinx.frontend.window.RaisedCosineWindower"/>
<component name="fft"
type="edu.cmu.sphinx.frontend.transform.DiscreteFourierTransform"/>
<component name="melFilterBank"
type="edu.cmu.sphinx.frontend.frequencywarp.MelFrequencyFilterBank"/>
<component name="dct"
type="edu.cmu.sphinx.frontend.transform.DiscreteCosineTransform"/>
<component name="liveCMN"
type="edu.cmu.sphinx.frontend.feature.LiveCMN"/>
<component name="featureExtraction"
type="edu.cmu.sphinx.frontend.feature.DeltasFeatureExtractor"/>
<!-- Newly Added.. -->
<component name="streamDataSource"
type="edu.cmu.sphinx.frontend.util.StreamDataSource">
<property name="sampleRate" value="16000"/>
<property name="bigEndianData" value="false"/>
</component>
<!-- ******************************************************* -->
<!-- monitors -->
<!-- ******************************************************* -->
<component name="accuracyTracker"
type="edu.cmu.sphinx.instrumentation.BestPathAccuracyTracker">
<property name="recognizer" value="${recognizer}"/>
<property name="showRawResults" value="false"/>
<property name="showAlignedResults" value="false"/>
</component>
<component name="memoryTracker"
type="edu.cmu.sphinx.instrumentation.MemoryTracker">
<property name="recognizer" value="${recognizer}"/>
<property name="showDetails" value="false"/>
<property name="showSummary" value="false"/>
</component>
<component name="speedTracker"
type="edu.cmu.sphinx.instrumentation.SpeedTracker">
<property name="recognizer" value="${recognizer}"/>
<property name="frontend" value="${frontend}"/>
<property name="showDetails" value="false"/>
</component>
<component name="recognizerMonitor"
type="edu.cmu.sphinx.instrumentation.RecognizerMonitor">
<property name="recognizer" value="${recognizer}"/>
<propertylist name="allocatedMonitors">
<item>configMonitor </item>
</propertylist>
</component>
<component name="configMonitor"
type="edu.cmu.sphinx.instrumentation.ConfigMonitor">
<property name="showConfig" value="false"/>
</component>
<!-- ******************************************************* -->
<!-- Miscellaneous components -->
<!-- ******************************************************* -->
<component name="logMath" type="edu.cmu.sphinx.util.LogMath">
<property name="logBase" value="1.0001"/>
<property name="useAddTable" value="true"/>
</component>
</config>
回答1:
The most common reasons for bad recognition accuracy are:
The mismatch of the sample rate of the incoming audio. It must be 16khz 16bit mono little-endian file. You need to fix sample rate of the source with resampling.
Zero silence regions in audio files decoded from mp3 break the decoder. You can use dither to introduce small random noise to solve this problem.
The mismatch of the acoustic model. You can use acoustic model adaptation to improve accuracy
The mismatch of the langauge model. You can create your own langauge model to match the vocabulary you are trying to decode.
You can get more information from the tutorial:
http://cmusphinx.sourceforge.net/wiki/tutorial
To get more detailed help you can always provide the audio samples you are trying to decode. They will help developers to analyze problem better. It's also helpful to provide the actual results you are getting from the decoder and your expectations.
回答2:
CMU Sphinx is working very good for me, just for the sake of sharing some knowledge, my setup is:
- Linux OS of course.
- I record 32kHz .wav files that I later pass to the Recognizer as the audioFileDataSource to get the Speech converted to Text.
- Trigram Language Model (SimpleNGramModel class)
- My Language Model is a custom one that I generated with the words/phrases I wanted. (Used the CMU Cam Toolkit version 2 (docs available at http://svr-www.eng.cam.ac.uk/~prc14/toolkit_documentation.html to generate my own trigram.arpa files)
- My Acoustic Model is wsj (TiedStateAcousticModel class) and wsjLoader (Sphinx3Loader class) with the WSJ_8gau_13dCep_8kHz_31mel_200Hz_3500Hz.jar (for some reason, this works better for me than the 16kHz model) and its dictionary.
- I use a Live FrontEnd with melFilterBank (tuned up to the acoustic model parameters) and liveCMN.
I think the key is to generate the appropriate trigram.arpa files using the tools.
You will have to tune up your sphinx config properties as needed, there is no magic bullet for that, some of the ones that helped me are the speechClassifierThreshold (44) and the speechMarkerTrailer (77).
Hope it helps or at least gives you some ideas.
来源：https://stackoverflow.com/questions/7411563/speech-recognition-response-is-poor-in-sphinx4