Наши проекты:
Журнал · Discuz!ML · Wiki · DRKB · Помощь проекту |
||
ПРАВИЛА | FAQ | Помощь | Поиск | Участники | Календарь | Избранное | RSS |
[3.15.5.183] |
|
Страницы: (23) « Первая ... 20 21 [22] 23 ( Перейти к последнему сообщению ) |
Сообщ.
#316
,
|
|
|
Предыдущий вопрос закрыт - была ошибка с моей стороны.
|
Сообщ.
#317
,
|
|
|
Привет
установил pocketsphinx и тестирую запись вот на этой модели https://sourceforge.net/projects/cmusphinx/...Models/Russian/ Качество как-то не очень. Можете подсказать где можно найти лучше модель? И в какую сторону копать, если нужно распознавать много разных голосов. |
Сообщ.
#318
,
|
|
|
Цитата dmarck @ Можете подсказать где можно найти лучше модель? Купить у Яндекса. |
Сообщ.
#319
,
|
|
|
Цитата nsh @ сейчас их и использую. Готов потерять в качестве, но прогонять больший объем для аналитики. |
Сообщ.
#320
,
|
|
|
Цитата dmarck @ сейчас их и использую. Готов потерять в качестве, но прогонять больший объем для аналитики. Тогда проще всего натренировать свою модель на результатах Яндекса. |
Сообщ.
#321
,
|
|
|
Добрый день, пытаюсь запустить sphinx на win7, gradle вроде строит нормально, но пока не удаётся запустить распознавание из файла тестовой записи на англ. 20 секунд голоса. Процесс вылетает из-за нехватки памяти, буду благодарен за подсказку
at java.io.FileInputStream.open0(Native Method) at java.io.FileInputStream.open(FileInputStream.java:195) at java.io.FileInputStream.<init>(FileInputStream.java:138) at java.io.FileInputStream.<init>(FileInputStream.java:93) at sun.net.www.protocol.file.FileURLConnection.connect(FileURLConnection .java:90) at sun.net.www.protocol.file.FileURLConnection.getInputStream(FileURLCon nection.java:188) at java.net.URL.openStream(URL.java:1045) at edu.cmu.sphinx.linguist.acoustic.tiedstate.Sphinx3Loader.getDataStrea m(Sphinx3Loader.java:233) at edu.cmu.sphinx.linguist.acoustic.tiedstate.Sphinx3Loader.readS3Binary Header(Sphinx3Loader.java:604) at edu.cmu.sphinx.linguist.acoustic.tiedstate.Sphinx3Loader.loadDensityF ile(Sphinx3Loader.java:525) at edu.cmu.sphinx.linguist.acoustic.tiedstate.Sphinx3Loader.loadModelFil es(Sphinx3Loader.java:298) at edu.cmu.sphinx.linguist.acoustic.tiedstate.Sphinx3Loader.load(Sphinx3 Loader.java:252) at edu.cmu.sphinx.frontend.AutoCepstrum.newProperties(AutoCepstrum.java: 118) ... 25 more FAILURE: Build failed with an exception. * What went wrong: Execution failed for task ':run'. > Process 'command 'C:\Program Files\Java\jdk1.8.0_144\bin\java.exe'' finished w ith non-zero exit value 1 * Try: Run with --stacktrace option to get the stack trace. Run with --info or --debug option to get more log output. * Get more help at https://help.gradle.org BUILD FAILED in 2s 2 actionable tasks: 1 executed, 1 up-to-date c:\workspace_Oxygen\Speech_Recognise>chcp ’ҐЄгй п Є®¤®ў п бва Ёж : 866 c:\workspace_Oxygen\Speech_Recognise>chcp 65001 Active code page: 65001 c:\workspace_Oxygen\Speech_Recognise>gradle run > Task :run Exception in thread "main" Property exception component:'null' property:'null' - java.io.FileNotFoundException: C:\workspace_Oxygen\Speech_Recognise\en-us\means (Íå óäàåòñÿ íàéòè óêàçàííûé ôàéë) edu.cmu.sphinx.util.props.PropertyException: java.io.FileNotFoundException: C:\w orkspace_Oxygen\Speech_Recognise\en-us\means (Íå óäàåòñÿ íàéòè óêàçàííûé ôàéë) at edu.cmu.sphinx.frontend.AutoCepstrum.newProperties(AutoCepstrum.java: 120) at edu.cmu.sphinx.util.props.PropertySheet.getOwner(PropertySheet.java:5 18) at edu.cmu.sphinx.util.props.ConfigurationManager.lookup(ConfigurationMa nager.java:163) at edu.cmu.sphinx.util.props.PropertySheet.getComponentList(PropertyShee t.java:432) at edu.cmu.sphinx.frontend.FrontEnd.newProperties(FrontEnd.java:160) at edu.cmu.sphinx.util.props.PropertySheet.getOwner(PropertySheet.java:5 18) at edu.cmu.sphinx.util.props.PropertySheet.getComponent(PropertySheet.ja va:296) at edu.cmu.sphinx.decoder.scorer.SimpleAcousticScorer.newProperties(Simp leAcousticScorer.java:47) at edu.cmu.sphinx.util.props.PropertySheet.getOwner(PropertySheet.java:5 18) at edu.cmu.sphinx.util.props.PropertySheet.getComponent(PropertySheet.ja va:296) at edu.cmu.sphinx.decoder.search.WordPruningBreadthFirstSearchManager.ne wProperties(WordPruningBreadthFirstSearchManager.java:213) at edu.cmu.sphinx.decoder.search.WordPruningBreadthFirstLookaheadSearchM anager.newProperties(WordPruningBreadthFirstLookaheadSearchManager.java:156) at edu.cmu.sphinx.util.props.PropertySheet.getOwner(PropertySheet.java:5 18) at edu.cmu.sphinx.util.props.PropertySheet.getComponent(PropertySheet.ja va:296) at edu.cmu.sphinx.decoder.AbstractDecoder.newProperties(AbstractDecoder. java:81) at edu.cmu.sphinx.decoder.Decoder.newProperties(Decoder.java:36) at edu.cmu.sphinx.util.props.PropertySheet.getOwner(PropertySheet.java:5 18) at edu.cmu.sphinx.util.props.PropertySheet.getComponent(PropertySheet.ja va:296) at edu.cmu.sphinx.recognizer.Recognizer.newProperties(Recognizer.java:86 ) at edu.cmu.sphinx.util.props.PropertySheet.getOwner(PropertySheet.java:5 18) at edu.cmu.sphinx.util.props.ConfigurationManager.lookup(ConfigurationMa nager.java:163) at edu.cmu.sphinx.api.Context.<init>(Context.java:73) at edu.cmu.sphinx.api.Context.<init>(Context.java:45) at edu.cmu.sphinx.api.AbstractSpeechRecognizer.<init>(AbstractSpeechReco gnizer.java:44) at edu.cmu.sphinx.api.StreamSpeechRecognizer.<init>(StreamSpeechRecogniz er.java:35) at Demo.TranscriberDemo.main(TranscriberDemo.java:21) Caused by: java.io.FileNotFoundException: C:\workspace_Oxygen\Speech_Recognise\e n-us\means (Íå óäàåòñÿ íàéòè óêàçàííûé ôàéë) at java.io.FileInputStream.open0(Native Method) at java.io.FileInputStream.open(FileInputStream.java:195) at java.io.FileInputStream.<init>(FileInputStream.java:138) at java.io.FileInputStream.<init>(FileInputStream.java:93) at sun.net.www.protocol.file.FileURLConnection.connect(FileURLConnection .java:90) at sun.net.www.protocol.file.FileURLConnection.getInputStream(FileURLCon nection.java:188) at java.net.URL.openStream(URL.java:1045) at edu.cmu.sphinx.linguist.acoustic.tiedstate.Sphinx3Loader.getDataStrea m(Sphinx3Loader.java:233) at edu.cmu.sphinx.linguist.acoustic.tiedstate.Sphinx3Loader.readS3Binary Header(Sphinx3Loader.java:604) at edu.cmu.sphinx.linguist.acoustic.tiedstate.Sphinx3Loader.loadDensityF ile(Sphinx3Loader.java:525) at edu.cmu.sphinx.linguist.acoustic.tiedstate.Sphinx3Loader.loadModelFil es(Sphinx3Loader.java:298) at edu.cmu.sphinx.linguist.acoustic.tiedstate.Sphinx3Loader.load(Sphinx3 Loader.java:252) at edu.cmu.sphinx.frontend.AutoCepstrum.newProperties(AutoCepstrum.java: 118) ... 25 more FAILURE: Build failed with an exception. * What went wrong: Execution failed for task ':run'. > Process 'command 'C:\Program Files\Java\jdk1.8.0_144\bin\java.exe'' finished w ith non-zero exit value 1 * Try: Run with --stacktrace option to get the stack trace. Run with --info or --debug option to get more log output. * Get more help at https://help.gradle.org BUILD FAILED in 2s 2 actionable tasks: 1 executed, 1 up-to-date c:\workspace_Oxygen\Speech_Recognise>gradle build BUILD SUCCESSFUL in 6s 5 actionable tasks: 4 executed, 1 up-to-date c:\workspace_Oxygen\Speech_Recognise>gradle run > Task :run 15:14:26.200 INFO unitManager CI Unit: *+NSN+ 15:14:26.200 INFO unitManager CI Unit: *+SPN+ 15:14:26.200 INFO unitManager CI Unit: AA 15:14:26.200 INFO unitManager CI Unit: AE 15:14:26.200 INFO unitManager CI Unit: AH 15:14:26.200 INFO unitManager CI Unit: AO 15:14:26.200 INFO unitManager CI Unit: AW 15:14:26.200 INFO unitManager CI Unit: AY 15:14:26.200 INFO unitManager CI Unit: B 15:14:26.215 INFO unitManager CI Unit: CH 15:14:26.215 INFO unitManager CI Unit: D 15:14:26.215 INFO unitManager CI Unit: DH 15:14:26.215 INFO unitManager CI Unit: EH 15:14:26.215 INFO unitManager CI Unit: ER 15:14:26.215 INFO unitManager CI Unit: EY 15:14:26.215 INFO unitManager CI Unit: F 15:14:26.215 INFO unitManager CI Unit: G 15:14:26.215 INFO unitManager CI Unit: HH 15:14:26.215 INFO unitManager CI Unit: IH 15:14:26.215 INFO unitManager CI Unit: IY 15:14:26.215 INFO unitManager CI Unit: JH 15:14:26.215 INFO unitManager CI Unit: K 15:14:26.215 INFO unitManager CI Unit: L 15:14:26.215 INFO unitManager CI Unit: M 15:14:26.215 INFO unitManager CI Unit: N 15:14:26.215 INFO unitManager CI Unit: NG 15:14:26.215 INFO unitManager CI Unit: OW 15:14:26.215 INFO unitManager CI Unit: OY 15:14:26.215 INFO unitManager CI Unit: P 15:14:26.215 INFO unitManager CI Unit: R 15:14:26.215 INFO unitManager CI Unit: S 15:14:26.215 INFO unitManager CI Unit: SH 15:14:26.215 INFO unitManager CI Unit: T 15:14:26.215 INFO unitManager CI Unit: TH 15:14:26.215 INFO unitManager CI Unit: UH 15:14:26.215 INFO unitManager CI Unit: UW 15:14:26.215 INFO unitManager CI Unit: V 15:14:26.215 INFO unitManager CI Unit: W 15:14:26.215 INFO unitManager CI Unit: Y 15:14:26.215 INFO unitManager CI Unit: Z 15:14:26.215 INFO unitManager CI Unit: ZH 15:14:29.230 INFO autoCepstrum Cepstrum component auto-configured as fol lows: autoCepstrum {MelFrequencyFilterBank, Denoise, DiscreteCosineTransform2, L ifter} 15:14:29.261 INFO dictionary Loading dictionary from: file:/C:/workspa ce_Oxygen/Speech_Recognise/en-us/cmudict-en-us.dict 15:14:29.901 INFO dictionary Loading filler dictionary from: file:/C:/ workspace_Oxygen/Speech_Recognise/en-us/en-us/noisedict 15:14:29.901 INFO acousticModelLoader Loading tied-state acoustic model from: f ile:/C:/workspace_Oxygen/Speech_Recognise/en-us/en-us 15:14:29.901 INFO acousticModelLoader Pool means Entries: 16128 15:14:29.901 INFO acousticModelLoader Pool variances Entries: 16128 15:14:29.901 INFO acousticModelLoader Pool transition_matrices Entries: 42 15:14:29.901 INFO acousticModelLoader Pool senones Entries: 5126 15:14:29.901 INFO acousticModelLoader Gaussian weights: mixture_weights. Entrie s: 15378 15:14:29.901 INFO acousticModelLoader Pool senones Entries: 5126 15:14:29.901 INFO acousticModelLoader Context Independent Unit Entries: 42 15:14:29.901 INFO acousticModelLoader HMM Manager: 137095 hmms 15:14:29.901 INFO acousticModel CompositeSenoneSequences: 0 15:14:29.901 INFO trieNgramModel Loading n-gram language model from: file: /C:/workspace_Oxygen/Speech_Recognise/en-us/en-us.lm.bin 15:14:43.667 INFO lexTreeLinguist Max CI Units 43 15:14:43.667 INFO lexTreeLinguist Unit table size 79507 15:14:43.683 INFO speedTracker # ----------------------------- Timers--- ------------------------------------- 15:14:43.683 INFO speedTracker # Name Count CurTime Mi nTime MaxTime AvgTime TotTime 15:14:43.683 INFO speedTracker Load AM 1 6,1520s 6, 1520s 6,1520s 6,1520s 6,1520s 15:14:43.683 INFO speedTracker Load Dictionary 1 0,6400s 0, 6400s 0,6400s 0,6400s 0,6400s 15:14:43.683 INFO speedTracker Load LM 1 11,1850s 11 ,1850s 11,1850s 11,1850s 11,1850s 15:14:43.683 INFO speedTracker Compile 1 2,5810s 2, 5810s 2,5810s 2,5810s 2,5810s 15:15:25.519 INFO liveCMN 55.31 -5.06 3.38 25.67 -3.44 -0.47 1.64 - 0.91 1.12 -3.75 2.71 -0.15 1.30 Exception in thread "main" java.lang.OutOfMemoryError: Java heap space at edu.cmu.sphinx.frontend.util.DataUtil.littleEndianBytesToValues(DataU til.java:176) at edu.cmu.sphinx.frontend.util.StreamDataSource.readNextFrame(StreamDat aSource.java:247) at edu.cmu.sphinx.frontend.util.StreamDataSource.getData(StreamDataSourc e.java:178) at edu.cmu.sphinx.frontend.DataBlocker.getData(DataBlocker.java:56) at edu.cmu.sphinx.frontend.endpoint.SpeechClassifier.getData(SpeechClass ifier.java:220) at edu.cmu.sphinx.frontend.endpoint.SpeechMarker.getData(SpeechMarker.ja va:136) at edu.cmu.sphinx.frontend.filter.Preemphasizer.getData(Preemphasizer.ja va:82) at edu.cmu.sphinx.frontend.window.RaisedCosineWindower.getData(RaisedCos ineWindower.java:154) at edu.cmu.sphinx.frontend.transform.DiscreteFourierTransform.getData(Di screteFourierTransform.java:269) at edu.cmu.sphinx.frontend.frequencywarp.MelFrequencyFilterBank.getData( MelFrequencyFilterBank.java:291) at edu.cmu.sphinx.frontend.denoise.Denoise.getData(Denoise.java:111) at edu.cmu.sphinx.frontend.transform.DiscreteCosineTransform.getData(Dis creteCosineTransform.java:77) at edu.cmu.sphinx.frontend.transform.Lifter.getData(Lifter.java:66) at edu.cmu.sphinx.frontend.AutoCepstrum.getData(AutoCepstrum.java:243) at edu.cmu.sphinx.frontend.feature.LiveCMN.getData(LiveCMN.java:177) at edu.cmu.sphinx.frontend.feature.AbstractFeatureExtractor.getNextData( AbstractFeatureExtractor.java:122) at edu.cmu.sphinx.frontend.feature.AbstractFeatureExtractor.getData(Abst ractFeatureExtractor.java:93) at edu.cmu.sphinx.frontend.feature.FeatureTransform.getData(FeatureTrans form.java:85) at edu.cmu.sphinx.frontend.FrontEnd.getData(FrontEnd.java:222) at edu.cmu.sphinx.decoder.scorer.SimpleAcousticScorer.getNextData(Simple AcousticScorer.java:146) at edu.cmu.sphinx.decoder.scorer.SimpleAcousticScorer.calculateScoresAnd StoreData(SimpleAcousticScorer.java:104) at edu.cmu.sphinx.decoder.search.WordPruningBreadthFirstLookaheadSearchM anager.scoreFastMatchTokens(WordPruningBreadthFirstLookaheadSearchManager.java:2 86) at edu.cmu.sphinx.decoder.search.WordPruningBreadthFirstLookaheadSearchM anager.fastMatchRecognize(WordPruningBreadthFirstLookaheadSearchManager.java:206 ) at edu.cmu.sphinx.decoder.search.WordPruningBreadthFirstLookaheadSearchM anager.recognize(WordPruningBreadthFirstLookaheadSearchManager.java:187) at edu.cmu.sphinx.decoder.Decoder.decode(Decoder.java:65) at edu.cmu.sphinx.recognizer.Recognizer.recognize(Recognizer.java:106) at edu.cmu.sphinx.recognizer.Recognizer.recognize(Recognizer.java:122) at edu.cmu.sphinx.api.AbstractSpeechRecognizer.getResult(AbstractSpeechR ecognizer.java:60) at Demo.TranscriberDemo.main(TranscriberDemo.java:26) FAILURE: Build failed with an exception. * What went wrong: Execution failed for task ':run'. > Process 'command 'C:\Program Files\Java\jdk1.8.0_144\bin\java.exe'' finished w ith non-zero exit value 1 * Try: Run with --stacktrace option to get the stack trace. Run with --info or --debug option to get more log output. * Get more help at https://help.gradle.org BUILD FAILED in 1m 58s 2 actionable tasks: 1 executed, 1 up-to-date c:\workspace_Oxygen\Speech_Recognise>^A c:\workspace_Oxygen\Speech_Recognise> Добавлено Собственно вот код проекта package Demo; import java.io.File; import java.io.FileInputStream; import java.io.InputStream; import edu.cmu.sphinx.api.Configuration; import edu.cmu.sphinx.api.SpeechResult; import edu.cmu.sphinx.api.StreamSpeechRecognizer; public class TranscriberDemo { public static void main(String[] args) throws Exception { Configuration configuration = new Configuration(); configuration.setAcousticModelPath("file:///C:/workspace_Oxygen/Speech_Recognise/en-us/en-us"); configuration.setDictionaryPath("file:///C:/workspace_Oxygen/Speech_Recognise/en-us/cmudict-en-us.dict"); configuration.setLanguageModelPath("file:///C:/workspace_Oxygen/Speech_Recognise/en-us/en-us.lm.bin"); StreamSpeechRecognizer recognizer = new StreamSpeechRecognizer(configuration); InputStream stream = new FileInputStream(new File("C:\\workspace_Oxygen\\Test_wav\\b0365.wav")); recognizer.startRecognition(stream); SpeechResult result; while ((result = recognizer.getResult()) != null) { System.out.format("Hypothesis: %s\n", result.getHypothesis()); } recognizer.stopRecognition(); } } |
Сообщ.
#322
,
|
|
|
Цитата Stanislav @ Памяти можно добавить с помощью ключа jvm -Xmx, например -Xmx2048m |
Сообщ.
#323
,
|
|
|
Например java -Xmx2048m
Так, изменил код немного package Demo; import java.io.File; import java.io.FileInputStream; import java.io.InputStream; import edu.cmu.sphinx.api.Configuration; import edu.cmu.sphinx.api.SpeechResult; import edu.cmu.sphinx.api.StreamSpeechRecognizer; public class TranscriberDemo { public static void main(String[] args) throws Exception { System.out.println ("+++++++++++++++++Work started1+++++++++++++++"); Configuration configuration = new Configuration(); configuration.setAcousticModelPath("file:///C:/workspace_Oxygen/Speech_Recognise/en-us/en-us"); configuration.setDictionaryPath("file:///C:/workspace_Oxygen/Speech_Recognise/en-us/cmudict-en-us.dict"); configuration.setLanguageModelPath("file:///C:/workspace_Oxygen/Speech_Recognise/en-us/en-us.lm.bin"); StreamSpeechRecognizer recognizer = new StreamSpeechRecognizer(configuration); InputStream stream = new FileInputStream(new File("C:\\workspace_Oxygen\\Test_wav\\b0365.wav")); System.out.println ("+++++++++++++++++Work started2+++++++++++++++"); recognizer.startRecognition(stream); System.out.println ("+++++++++++++++++Work started3+++++++++++++++"); SpeechResult result; System.out.println ("+++++++++++++++++Work resulted complete1+++++++++++++++"); while ((result = recognizer.getResult()) != null) { System.out.println ("+++++++++++++++++recognizer.getResult+++++++++++++++"); System.out.format("Hypothesis: %s\n", result.getHypothesis()); System.out.println ("^^^^^^^^^^^^^^"+result.getHypothesis()+"^^^^^^^^^^^^^^^^^^^^^"); } recognizer.stopRecognition(); } } Результат работы так же вылетает по памяти, но как видно в конце, при получении результатов 16:11:07.734 INFO unitManager CI Unit: CH 16:11:07.735 INFO unitManager CI Unit: D 16:11:07.735 INFO unitManager CI Unit: DH 16:11:07.735 INFO unitManager CI Unit: EH 16:11:07.735 INFO unitManager CI Unit: ER 16:11:07.736 INFO unitManager CI Unit: EY 16:11:07.736 INFO unitManager CI Unit: F 16:11:07.736 INFO unitManager CI Unit: G 16:11:07.737 INFO unitManager CI Unit: HH 16:11:07.737 INFO unitManager CI Unit: IH 16:11:07.737 INFO unitManager CI Unit: IY 16:11:07.737 INFO unitManager CI Unit: JH 16:11:07.738 INFO unitManager CI Unit: K 16:11:07.738 INFO unitManager CI Unit: L 16:11:07.738 INFO unitManager CI Unit: M 16:11:07.738 INFO unitManager CI Unit: N 16:11:07.738 INFO unitManager CI Unit: NG 16:11:07.739 INFO unitManager CI Unit: OW 16:11:07.739 INFO unitManager CI Unit: OY 16:11:07.739 INFO unitManager CI Unit: P 16:11:07.740 INFO unitManager CI Unit: R 16:11:07.740 INFO unitManager CI Unit: S 16:11:07.740 INFO unitManager CI Unit: SH 16:11:07.740 INFO unitManager CI Unit: T 16:11:07.741 INFO unitManager CI Unit: TH 16:11:07.741 INFO unitManager CI Unit: UH 16:11:07.741 INFO unitManager CI Unit: UW 16:11:07.742 INFO unitManager CI Unit: V 16:11:07.742 INFO unitManager CI Unit: W 16:11:07.742 INFO unitManager CI Unit: Y 16:11:07.742 INFO unitManager CI Unit: Z 16:11:07.743 INFO unitManager CI Unit: ZH 16:11:11.204 INFO autoCepstrum Cepstrum component auto-configured as fol lows: autoCepstrum {MelFrequencyFilterBank, Denoise, DiscreteCosineTransform2, L ifter} +++++++++++++++++Work started2+++++++++++++++ 16:11:11.231 INFO dictionary Loading dictionary from: file:/C:/workspa ce_Oxygen/Speech_Recognise/en-us/cmudict-en-us.dict 16:11:11.988 INFO dictionary Loading filler dictionary from: file:/C:/ workspace_Oxygen/Speech_Recognise/en-us/en-us/noisedict 16:11:11.988 INFO acousticModelLoader Loading tied-state acoustic model from: f ile:/C:/workspace_Oxygen/Speech_Recognise/en-us/en-us 16:11:11.988 INFO acousticModelLoader Pool means Entries: 16128 16:11:11.988 INFO acousticModelLoader Pool variances Entries: 16128 16:11:11.988 INFO acousticModelLoader Pool transition_matrices Entries: 42 16:11:11.988 INFO acousticModelLoader Pool senones Entries: 5126 16:11:11.988 INFO acousticModelLoader Gaussian weights: mixture_weights. Entrie s: 15378 16:11:11.988 INFO acousticModelLoader Pool senones Entries: 5126 16:11:11.988 INFO acousticModelLoader Context Independent Unit Entries: 42 16:11:11.988 INFO acousticModelLoader HMM Manager: 137095 hmms 16:11:11.988 INFO acousticModel CompositeSenoneSequences: 0 16:11:11.988 INFO trieNgramModel Loading n-gram language model from: file: /C:/workspace_Oxygen/Speech_Recognise/en-us/en-us.lm.bin 16:11:27.155 INFO lexTreeLinguist Max CI Units 43 16:11:27.155 INFO lexTreeLinguist Unit table size 79507 16:11:27.155 INFO speedTracker # ----------------------------- Timers--- ------------------------------------- 16:11:27.155 INFO speedTracker # Name Count CurTime Mi nTime MaxTime AvgTime TotTime 16:11:27.155 INFO speedTracker Load AM 1 7,0880s 7, 0880s 7,0880s 7,0880s 7,0880s 16:11:27.155 INFO speedTracker Load Dictionary 1 0,7570s 0, 7570s 0,7570s 0,7570s 0,7570s 16:11:27.155 INFO speedTracker Load LM 1 12,2460s 12 ,2460s 12,2460s 12,2460s 12,2460s 16:11:27.171 INFO speedTracker Compile 1 2,9200s 2, 9200s 2,9200s 2,9200s 2,9200s +++++++++++++++++Work started3+++++++++++++++ +++++++++++++++++Work resulted complete1+++++++++++++++ 16:12:15.438 INFO liveCMN 55.31 -5.06 3.38 25.67 -3.44 -0.47 1.64 - 0.91 1.12 -3.75 2.71 -0.15 1.30 Exception in thread "main" java.lang.OutOfMemoryError: Java heap space at edu.cmu.sphinx.frontend.util.DataUtil.littleEndianBytesToValues(DataU til.java:176) at edu.cmu.sphinx.frontend.util.StreamDataSource.readNextFrame(StreamDat aSource.java:247) at edu.cmu.sphinx.frontend.util.StreamDataSource.getData(StreamDataSourc e.java:178) at edu.cmu.sphinx.frontend.DataBlocker.getData(DataBlocker.java:56) at edu.cmu.sphinx.frontend.endpoint.SpeechClassifier.getData(SpeechClass ifier.java:220) at edu.cmu.sphinx.frontend.endpoint.SpeechMarker.getData(SpeechMarker.ja va:136) at edu.cmu.sphinx.frontend.filter.Preemphasizer.getData(Preemphasizer.ja va:82) at edu.cmu.sphinx.frontend.window.RaisedCosineWindower.getData(RaisedCos ineWindower.java:154) at edu.cmu.sphinx.frontend.transform.DiscreteFourierTransform.getData(Di screteFourierTransform.java:269) at edu.cmu.sphinx.frontend.frequencywarp.MelFrequencyFilterBank.getData( MelFrequencyFilterBank.java:291) at edu.cmu.sphinx.frontend.denoise.Denoise.getData(Denoise.java:111) at edu.cmu.sphinx.frontend.transform.DiscreteCosineTransform.getData(Dis creteCosineTransform.java:77) at edu.cmu.sphinx.frontend.transform.Lifter.getData(Lifter.java:66) at edu.cmu.sphinx.frontend.AutoCepstrum.getData(AutoCepstrum.java:243) at edu.cmu.sphinx.frontend.feature.LiveCMN.getData(LiveCMN.java:177) at edu.cmu.sphinx.frontend.feature.AbstractFeatureExtractor.getNextData( AbstractFeatureExtractor.java:122) at edu.cmu.sphinx.frontend.feature.AbstractFeatureExtractor.getData(Abst ractFeatureExtractor.java:93) at edu.cmu.sphinx.frontend.feature.FeatureTransform.getData(FeatureTrans form.java:85) at edu.cmu.sphinx.frontend.FrontEnd.getData(FrontEnd.java:222) at edu.cmu.sphinx.decoder.scorer.SimpleAcousticScorer.getNextData(Simple AcousticScorer.java:146) at edu.cmu.sphinx.decoder.scorer.SimpleAcousticScorer.calculateScoresAnd StoreData(SimpleAcousticScorer.java:104) at edu.cmu.sphinx.decoder.search.WordPruningBreadthFirstLookaheadSearchM anager.scoreFastMatchTokens(WordPruningBreadthFirstLookaheadSearchManager.java:2 86) at edu.cmu.sphinx.decoder.search.WordPruningBreadthFirstLookaheadSearchM anager.fastMatchRecognize(WordPruningBreadthFirstLookaheadSearchManager.java:206 ) at edu.cmu.sphinx.decoder.search.WordPruningBreadthFirstLookaheadSearchM anager.recognize(WordPruningBreadthFirstLookaheadSearchManager.java:187) at edu.cmu.sphinx.decoder.Decoder.decode(Decoder.java:65) at edu.cmu.sphinx.recognizer.Recognizer.recognize(Recognizer.java:106) at edu.cmu.sphinx.recognizer.Recognizer.recognize(Recognizer.java:122) at edu.cmu.sphinx.api.AbstractSpeechRecognizer.getResult(AbstractSpeechR ecognizer.java:60) at Demo.TranscriberDemo.main(TranscriberDemo.java:28) FAILURE: Build failed with an exception. * What went wrong: Execution failed for task ':run'. > Process 'command 'C:\Program Files\Java\jdk1.8.0_144\bin\java.exe'' finished w ith non-zero exit value 1 * Try: Run with --stacktrace option to get the stack trace. Run with --info or --debug option to get more log output. * Get more help at https://help.gradle.org BUILD FAILED in 2m 18s 2 actionable tasks: 1 executed, 1 up-to-date c:\workspace_Oxygen\Speech_Recognise>gradle build BUILD SUCCESSFUL in 7s 5 actionable tasks: 4 executed, 1 up-to-date c:\workspace_Oxygen\Speech_Recognise>gradle run > Task :run +++++++++++++++++Work started1+++++++++++++++ 16:16:13.292 INFO unitManager CI Unit: *+NSN+ 16:16:13.298 INFO unitManager CI Unit: *+SPN+ 16:16:13.299 INFO unitManager CI Unit: AA 16:16:13.299 INFO unitManager CI Unit: AE 16:16:13.299 INFO unitManager CI Unit: AH 16:16:13.300 INFO unitManager CI Unit: AO 16:16:13.300 INFO unitManager CI Unit: AW 16:16:13.300 INFO unitManager CI Unit: AY 16:16:13.302 INFO unitManager CI Unit: B 16:16:13.302 INFO unitManager CI Unit: CH 16:16:13.303 INFO unitManager CI Unit: D 16:16:13.304 INFO unitManager CI Unit: DH 16:16:13.304 INFO unitManager CI Unit: EH 16:16:13.304 INFO unitManager CI Unit: ER 16:16:13.304 INFO unitManager CI Unit: EY 16:16:13.304 INFO unitManager CI Unit: F 16:16:13.305 INFO unitManager CI Unit: G 16:16:13.305 INFO unitManager CI Unit: HH 16:16:13.305 INFO unitManager CI Unit: IH 16:16:13.305 INFO unitManager CI Unit: IY 16:16:13.306 INFO unitManager CI Unit: JH 16:16:13.306 INFO unitManager CI Unit: K 16:16:13.306 INFO unitManager CI Unit: L 16:16:13.306 INFO unitManager CI Unit: M 16:16:13.306 INFO unitManager CI Unit: N 16:16:13.307 INFO unitManager CI Unit: NG 16:16:13.307 INFO unitManager CI Unit: OW 16:16:13.307 INFO unitManager CI Unit: OY 16:16:13.307 INFO unitManager CI Unit: P 16:16:13.307 INFO unitManager CI Unit: R 16:16:13.307 INFO unitManager CI Unit: S 16:16:13.308 INFO unitManager CI Unit: SH 16:16:13.308 INFO unitManager CI Unit: T 16:16:13.308 INFO unitManager CI Unit: TH 16:16:13.308 INFO unitManager CI Unit: UH 16:16:13.308 INFO unitManager CI Unit: UW 16:16:13.309 INFO unitManager CI Unit: V 16:16:13.310 INFO unitManager CI Unit: W 16:16:13.310 INFO unitManager CI Unit: Y 16:16:13.310 INFO unitManager CI Unit: Z 16:16:13.310 INFO unitManager CI Unit: ZH 16:16:16.654 INFO autoCepstrum Cepstrum component auto-configured as fol lows: autoCepstrum {MelFrequencyFilterBank, Denoise, DiscreteCosineTransform2, L ifter} +++++++++++++++++Work started2+++++++++++++++ 16:16:16.661 INFO dictionary Loading dictionary from: file:/C:/workspa ce_Oxygen/Speech_Recognise/en-us/cmudict-en-us.dict 16:16:17.331 INFO dictionary Loading filler dictionary from: file:/C:/ workspace_Oxygen/Speech_Recognise/en-us/en-us/noisedict 16:16:17.332 INFO acousticModelLoader Loading tied-state acoustic model from: f ile:/C:/workspace_Oxygen/Speech_Recognise/en-us/en-us 16:16:17.333 INFO acousticModelLoader Pool means Entries: 16128 16:16:17.333 INFO acousticModelLoader Pool variances Entries: 16128 16:16:17.333 INFO acousticModelLoader Pool transition_matrices Entries: 42 16:16:17.333 INFO acousticModelLoader Pool senones Entries: 5126 16:16:17.334 INFO acousticModelLoader Gaussian weights: mixture_weights. Entrie s: 15378 16:16:17.334 INFO acousticModelLoader Pool senones Entries: 5126 16:16:17.334 INFO acousticModelLoader Context Independent Unit Entries: 42 16:16:17.335 INFO acousticModelLoader HMM Manager: 137095 hmms 16:16:17.335 INFO acousticModel CompositeSenoneSequences: 0 16:16:17.336 INFO trieNgramModel Loading n-gram language model from: file: /C:/workspace_Oxygen/Speech_Recognise/en-us/en-us.lm.bin 16:16:31.195 INFO lexTreeLinguist Max CI Units 43 16:16:31.196 INFO lexTreeLinguist Unit table size 79507 16:16:31.200 INFO speedTracker # ----------------------------- Timers--- ------------------------------------- 16:16:31.200 INFO speedTracker # Name Count CurTime Mi nTime MaxTime AvgTime TotTime 16:16:31.201 INFO speedTracker Load AM 1 6,6330s 6, 6330s 6,6330s 6,6330s 6,6330s 16:16:31.202 INFO speedTracker Load Dictionary 1 0,6710s 0, 6710s 0,6710s 0,6710s 0,6710s 16:16:31.202 INFO speedTracker Load LM 1 10,9540s 10 ,9540s 10,9540s 10,9540s 10,9540s 16:16:31.202 INFO speedTracker Compile 1 2,9070s 2, 9070s 2,9070s 2,9070s 2,9070s +++++++++++++++++Work started3+++++++++++++++ +++++++++++++++++Work resulted complete1+++++++++++++++ 16:17:13.902 INFO liveCMN 55.31 -5.06 3.38 25.67 -3.44 -0.47 1.64 - 0.91 1.12 -3.75 2.71 -0.15 1.30 Exception in thread "main" java.lang.OutOfMemoryError: Java heap space at edu.cmu.sphinx.frontend.util.DataUtil.littleEndianBytesToValues(DataU til.java:176) at edu.cmu.sphinx.frontend.util.StreamDataSource.readNextFrame(StreamDat aSource.java:247) at edu.cmu.sphinx.frontend.util.StreamDataSource.getData(StreamDataSourc e.java:178) at edu.cmu.sphinx.frontend.DataBlocker.getData(DataBlocker.java:56) at edu.cmu.sphinx.frontend.endpoint.SpeechClassifier.getData(SpeechClass ifier.java:220) at edu.cmu.sphinx.frontend.endpoint.SpeechMarker.getData(SpeechMarker.ja va:136) at edu.cmu.sphinx.frontend.filter.Preemphasizer.getData(Preemphasizer.ja va:82) at edu.cmu.sphinx.frontend.window.RaisedCosineWindower.getData(RaisedCos ineWindower.java:154) at edu.cmu.sphinx.frontend.transform.DiscreteFourierTransform.getData(Di screteFourierTransform.java:269) at edu.cmu.sphinx.frontend.frequencywarp.MelFrequencyFilterBank.getData( MelFrequencyFilterBank.java:291) at edu.cmu.sphinx.frontend.denoise.Denoise.getData(Denoise.java:111) at edu.cmu.sphinx.frontend.transform.DiscreteCosineTransform.getData(Dis creteCosineTransform.java:77) at edu.cmu.sphinx.frontend.transform.Lifter.getData(Lifter.java:66) at edu.cmu.sphinx.frontend.AutoCepstrum.getData(AutoCepstrum.java:243) at edu.cmu.sphinx.frontend.feature.LiveCMN.getData(LiveCMN.java:177) at edu.cmu.sphinx.frontend.feature.AbstractFeatureExtractor.getNextData( AbstractFeatureExtractor.java:122) at edu.cmu.sphinx.frontend.feature.AbstractFeatureExtractor.getData(Abst ractFeatureExtractor.java:93) at edu.cmu.sphinx.frontend.feature.FeatureTransform.getData(FeatureTrans form.java:85) at edu.cmu.sphinx.frontend.FrontEnd.getData(FrontEnd.java:222) at edu.cmu.sphinx.decoder.scorer.SimpleAcousticScorer.getNextData(Simple AcousticScorer.java:146) at edu.cmu.sphinx.decoder.scorer.SimpleAcousticScorer.calculateScoresAnd StoreData(SimpleAcousticScorer.java:104) at edu.cmu.sphinx.decoder.search.WordPruningBreadthFirstLookaheadSearchM anager.scoreFastMatchTokens(WordPruningBreadthFirstLookaheadSearchManager.java:2 86) at edu.cmu.sphinx.decoder.search.WordPruningBreadthFirstLookaheadSearchM anager.fastMatchRecognize(WordPruningBreadthFirstLookaheadSearchManager.java:206 ) at edu.cmu.sphinx.decoder.search.WordPruningBreadthFirstLookaheadSearchM anager.recognize(WordPruningBreadthFirstLookaheadSearchManager.java:187) at edu.cmu.sphinx.decoder.Decoder.decode(Decoder.java:65) at edu.cmu.sphinx.recognizer.Recognizer.recognize(Recognizer.java:106) at edu.cmu.sphinx.recognizer.Recognizer.recognize(Recognizer.java:122) at edu.cmu.sphinx.api.AbstractSpeechRecognizer.getResult(AbstractSpeechR ecognizer.java:60) at Demo.TranscriberDemo.main(TranscriberDemo.java:28) FAILURE: Build failed with an exception. * What went wrong: Execution failed for task ':run'. > Process 'command 'C:\Program Files\Java\jdk1.8.0_144\bin\java.exe'' finished w ith non-zero exit value 1 * Try: Run with --stacktrace option to get the stack trace. Run with --info or --debug option to get more log output. * Get more help at https://help.gradle.org BUILD FAILED in 2m 5s 2 actionable tasks: 1 executed, 1 up-to-date c:\workspace_Oxygen\Speech_Recognise> То есть последнее выведённое сообщение +++++++++++++++++Work resulted complete1+++++++++++++++ Такие дела, нижайше кланяюсь и прошу помощи. |
Сообщ.
#324
,
|
|
|
Цитата Stanislav @ Например java -Xmx2048m Где например, что например? Ничего не поменялось, как я вижу. Цитата Такие дела, нижайше кланяюсь и прошу помощи. С мышлением раба далеко не уйдешь. Вместо всего этого лучше документацию почитать вдумчиво. |
Сообщ.
#325
,
|
|
|
Всех приветствую! Использую pocketsphinx для приложения "умный дом" на Android, необходимо реализовать голосовое управление. Для команд написал JSGF файлик следующего содержания (условно):
#JSGF V1.0; grammar words; public <words> = включи | выключи | свет; Для тестов проигрываю записи "включи свет" и "выключи свет". Однако, сфинкс не определяет всю фразу целиком, а выдает только "включи" или "выключи". Если добавить в JSGF полные фразы, начинает работать. Правильно ли я понимаю, что так и должно работать? Что сфинкс не может комбинировать слова в фразы из JSGF? Возможно, есть какое-то решение данного вопроса? Дело в том, что в будущем могут добавиться такие команды как "включи кондиционер" и так далее, неужели придется описывать все возможные комбинации? |
Сообщ.
#326
,
|
|
|
Цитата #JSGF V1.0; grammar words; public <words> = ( включи | выключи | свет )+; А если так? |
Сообщ.
#327
,
|
|
|
zamir спасибо, помогло! Есть какие-нибудь материалы по синтаксису JSGF?
|
Сообщ.
#328
,
|
|
|
Нет проблем
Информации в интернет было достаточно, когда я искал. Вот вторая ссылка в яндекс например: https://www.w3.org/TR/jsgf/ |
Сообщ.
#329
,
|
|
|
Цитата nsh @ Цитата Stanislav @ Памяти можно добавить с помощью ключа jvm -Xmx, например -Xmx2048m Чтоб тебе так помогали... Operation Format: -Xmx<size>[g|G|m|M|k|K] Combine -Xmx with a memory value For Example: java -Xmx:1g myApp sets the maximum java heap to 1 gigabyte. If you do not add a unit, you will get the exact value you state; for example, 64 will be interpreted as 64 bytes, not 64 megabytes or 64 kilobytes. The -Xmx option and -Xms option in combination are used to limit the Java heap size. The Java heap can never grow larger than -Xmx. Also, the -Xms value can be used as “minimum heap size” to set a fixed heap size by setting -Xms = -Xmx when, for example, you want to run benchmark tests. Собственно и в прочих "умных" мануалах то опечатка, то недосказанность, а в итоге даже пример из документации от sphinx не работает |
Сообщ.
#330
,
|
|
|
Так, для изменения памяти при запуске программы через gradle run годится вариант:
1. Создать файл gradle.properties рядом с gradle.build и 2. прописать там строчку "org.gradle.jvmargs=-Xms1024m -Xmx1024m" более подробная инфа в мануале по gradle И вопрос: какие минимальные требования к железу для нормальной работы? Вот у меня 1,5 Гб озу и файл подкачки стоит, но вылетает из-за нехватки памяти якобы, это нормальная работа программы или в коде есть ошибка? |