- This topic has 12 replies, 3 voices, and was last updated 8 years, 5 months ago by Halle Winkler.
-
AuthorPosts
-
November 1, 2015 at 2:25 am #1027141lytedesignsParticipant
Hello, we are having problems using OpenEars in Spanish.
The main problem is that it detects words have not said (sometimes a single hit in a table can generate a word).
We tested with Rejecto to cancel the words not in the dictionary, but problems remain.
All this, testing it in your main example project.(In our project we also use the plugin RapidEar for live detect, but work similarly. I mention this for what it’s worth )
We tried in Iphone 6, Iphone 4s, and all works just.
We have tried to generate the test file with SaveTheWave or ‘afconvert’ (as you say in the documentation) for the tests and any options gives good results (Where more problems given is with real rec.).
All it tested in a silent environment.
We also tried to change the ‘vadThreshold’ to ‘4.3’ as we have read in some post, but nothing changes.Another important thing is that the ‘recognitionScore’ value seems almost random.
For example: if we say “DERECHA”, and the app is thought to have said “CABEZA”, the result of ‘recognitionScore’ changes not just about whether we would have that right.We need help. Thanks
The last test we performed:1. We downloaded the latest version of OpenEars with its own example.
2. We added the framework Rejecto to the project.
3. In the example we have changed the OEAcousticModel to Spanish and the first array with Spanish words.
4. We test with wav file (added to project), and with the voice in real time
5. IMPORTANT: The words are not recognized well, or when they are not recognized.
6. We have the console log.Code changes in the openears example project and the logs of console are explained below:
WAV test:
——————
https://dl.dropboxusercontent.com/u/87410097/Rec2.wavCHANGE IMPORTS TO ADD REJECTO
————-
//#import <OpenEars/OELanguageModelGenerator.h>
#import <RejectoDemo/OELanguageModelGenerator+Rejecto.h>NEW WORDS TO DETECT
————-
NSArray *firstLanguageArray = @[@”CABEZA”,
@”CASA”,
@”DERECHA”,
@”SOBRE”,
@”HOLA”];THRESOLD CHANGE
————-
[OEPocketsphinxController sharedInstance].vadThreshold = 4.3f;CHANGE languageModelGenerator METHOD TO REJECTO
————-
//NSError *error = [languageModelGenerator generateLanguageModelFromArray:firstLanguageArray withFilesNamed:@”FirstOpenEarsDynamicLanguageModel” forAcousticModelAtPath:[OEAcousticModel pathToModel:@”AcousticModelEnglish”]];
//METHOD 2 – Rejecto
NSError *error = [languageModelGenerator generateRejectingLanguageModelFromArray:firstLanguageArray
withFilesNamed:@”FirstOpenEarsDynamicLanguageModel”
withOptionalExclusions:nil
usingVowelsOnly:FALSE
withWeight:nil
forAcousticModelAtPath:[OEAcousticModel pathToModel:@”AcousticModelSpanish”]];CONSOLE OUTPUT
————-
[spoiler]
2015-11-01 00:42:22.068 OpenEarsSampleApp[2568:1401541] Starting OpenEars logging for OpenEars version 2.041 on 64-bit device (or build): iPhone running iOS version: 9.000000
2015-11-01 00:42:22.069 OpenEarsSampleApp[2568:1401541] Creating shared instance of OEPocketsphinxController
2015-11-01 00:42:22.130 OpenEarsSampleApp[2568:1401541] I’m done running performDictionaryLookup and it took 0.003569 seconds
2015-11-01 00:42:22.131 OpenEarsSampleApp[2568:1401541] I’m done running performDictionaryLookup and it took 0.015418 seconds
2015-11-01 00:42:22.138 OpenEarsSampleApp[2568:1401541] Starting dynamic language model generationINFO: cmd_ln.c(703): Parsing command line:
sphinx_lm_convert \
-i /var/mobile/Containers/Data/Application/6E796B17-868E-448C-8B0F-093B1FF028E8/Library/Caches/FirstOpenEarsDynamicLanguageModel.arpa \
-o /var/mobile/Containers/Data/Application/6E796B17-868E-448C-8B0F-093B1FF028E8/Library/Caches/FirstOpenEarsDynamicLanguageModel.DMPCurrent configuration:
[NAME] [DEFLT] [VALUE]
-case
-debug 0
-help no no
-i /var/mobile/Containers/Data/Application/6E796B17-868E-448C-8B0F-093B1FF028E8/Library/Caches/FirstOpenEarsDynamicLanguageModel.arpa
-ifmt
-logbase 1.0001 1.000100e+00
-mmap no no
-o /var/mobile/Containers/Data/Application/6E796B17-868E-448C-8B0F-093B1FF028E8/Library/Caches/FirstOpenEarsDynamicLanguageModel.DMP
-ofmtINFO: ngram_model_arpa.c(503): ngrams 1=33, 2=62, 3=31
INFO: ngram_model_arpa.c(135): Reading unigrams
INFO: ngram_model_arpa.c(542): 33 = #unigrams created
INFO: ngram_model_arpa.c(195): Reading bigrams
INFO: ngram_model_arpa.c(560): 62 = #bigrams created
INFO: ngram_model_arpa.c(561): 3 = #prob2 entries
INFO: ngram_model_arpa.c(569): 3 = #bo_wt2 entries
INFO: ngram_model_arpa.c(292): Reading trigrams
INFO: ngram_model_arpa.c(582): 31 = #trigrams created
INFO: ngram_model_arpa.c(583): 2 = #prob3 entries
INFO: ngram_model_dmp.c(518): Building DMP model…
INFO: ngram_model_dmp.c(548): 33 = #unigrams created
INFO: ngram_model_dmp.c(649): 62 = #bigrams created
INFO: ngram_model_dmp.c(650): 3 = #prob2 entries
INFO: ngram_model_dmp.c(657): 3 = #bo_wt2 entries
INFO: ngram_model_dmp.c(661): 31 = #trigrams created
INFO: ngram_model_dmp.c(662): 2 = #prob3 entries
2015-11-01 00:42:22.186 OpenEarsSampleApp[2568:1401541] Done creating language model with CMUCLMTK in 0.048384 seconds.
INFO: cmd_ln.c(703): Parsing command line:
sphinx_lm_convert \
-i /var/mobile/Containers/Data/Application/6E796B17-868E-448C-8B0F-093B1FF028E8/Library/Caches/FirstOpenEarsDynamicLanguageModel.arpa \
-o /var/mobile/Containers/Data/Application/6E796B17-868E-448C-8B0F-093B1FF028E8/Library/Caches/FirstOpenEarsDynamicLanguageModel.DMPCurrent configuration:
[NAME] [DEFLT] [VALUE]
-case
-debug 0
-help no no
-i /var/mobile/Containers/Data/Application/6E796B17-868E-448C-8B0F-093B1FF028E8/Library/Caches/FirstOpenEarsDynamicLanguageModel.arpa
-ifmt
-logbase 1.0001 1.000100e+00
-mmap no no
-o /var/mobile/Containers/Data/Application/6E796B17-868E-448C-8B0F-093B1FF028E8/Library/Caches/FirstOpenEarsDynamicLanguageModel.DMP
-ofmtINFO: ngram_model_arpa.c(503): ngrams 1=33, 2=62, 3=31
INFO: ngram_model_arpa.c(135): Reading unigrams
INFO: ngram_model_arpa.c(542): 33 = #unigrams created
INFO: ngram_model_arpa.c(195): Reading bigrams
INFO: ngram_model_arpa.c(560): 62 = #bigrams created
INFO: ngram_model_arpa.c(561): 3 = #prob2 entries
INFO: ngram_model_arpa.c(569): 3 = #bo_wt2 entries
INFO: ngram_model_arpa.c(292): Reading trigrams
INFO: ngram_model_arpa.c(582): 31 = #trigrams created
INFO: ngram_model_arpa.c(583): 2 = #prob3 entries
INFO: ngram_model_dmp.c(518): Building DMP model…
INFO: ngram_model_dmp.c(548): 33 = #unigrams created
INFO: ngram_model_dmp.c(649): 62 = #bigrams created
INFO: ngram_model_dmp.c(650): 3 = #prob2 entries
INFO: ngram_model_dmp.c(657): 3 = #bo_wt2 entries
INFO: ngram_model_dmp.c(661): 31 = #trigrams created
INFO: ngram_model_dmp.c(662): 2 = #prob3 entries
2015-11-01 00:42:22.193 OpenEarsSampleApp[2568:1401541] I’m done running dynamic language model generation and it took 0.119459 seconds
2015-11-01 00:42:22.244 OpenEarsSampleApp[2568:1401541] The word QUIDNUNC was not found in the dictionary /var/mobile/Containers/Bundle/Application/CDC70BDD-C0A4-42C6-9B69-6E89E5B4B42F/OpenEarsSampleApp.app/AcousticModelEnglish.bundle/LanguageModelGeneratorLookupList.text/LanguageModelGeneratorLookupList.text.
2015-11-01 00:42:22.244 OpenEarsSampleApp[2568:1401541] Now using the fallback method to look up the word QUIDNUNC
2015-11-01 00:42:22.244 OpenEarsSampleApp[2568:1401541] If this is happening more frequently than you would expect, the most likely cause for it is since you are using the English phonetic lookup dictionary is that your words are not in English or aren’t dictionary words.
2015-11-01 00:42:22.244 OpenEarsSampleApp[2568:1401541] Using convertGraphemes for the word or phrase QUIDNUNC which doesn’t appear in the dictionary
2015-11-01 00:42:22.260 OpenEarsSampleApp[2568:1401541] I’m done running performDictionaryLookup and it took 0.048007 seconds
2015-11-01 00:42:22.262 OpenEarsSampleApp[2568:1401541] I’m done running performDictionaryLookup and it took 0.053857 seconds
2015-11-01 00:42:22.274 OpenEarsSampleApp[2568:1401541] Starting dynamic language model generationINFO: cmd_ln.c(703): Parsing command line:
sphinx_lm_convert \
-i /var/mobile/Containers/Data/Application/6E796B17-868E-448C-8B0F-093B1FF028E8/Library/Caches/SecondOpenEarsDynamicLanguageModel.arpa \
-o /var/mobile/Containers/Data/Application/6E796B17-868E-448C-8B0F-093B1FF028E8/Library/Caches/SecondOpenEarsDynamicLanguageModel.DMPCurrent configuration:
[NAME] [DEFLT] [VALUE]
-case
-debug 0
-help no no
-i /var/mobile/Containers/Data/Application/6E796B17-868E-448C-8B0F-093B1FF028E8/Library/Caches/SecondOpenEarsDynamicLanguageModel.arpa
-ifmt
-logbase 1.0001 1.000100e+00
-mmap no no
-o /var/mobile/Containers/Data/Application/6E796B17-868E-448C-8B0F-093B1FF028E8/Library/Caches/SecondOpenEarsDynamicLanguageModel.DMP
-ofmtINFO: ngram_model_arpa.c(503): ngrams 1=51, 2=97, 3=49
INFO: ngram_model_arpa.c(135): Reading unigrams
INFO: ngram_model_arpa.c(542): 51 = #unigrams created
INFO: ngram_model_arpa.c(195): Reading bigrams
INFO: ngram_model_arpa.c(560): 97 = #bigrams created
INFO: ngram_model_arpa.c(561): 3 = #prob2 entries
INFO: ngram_model_arpa.c(569): 3 = #bo_wt2 entries
INFO: ngram_model_arpa.c(292): Reading trigrams
INFO: ngram_model_arpa.c(582): 49 = #trigrams created
INFO: ngram_model_arpa.c(583): 2 = #prob3 entries
INFO: ngram_model_dmp.c(518): Building DMP model…
INFO: ngram_model_dmp.c(548): 51 = #unigrams created
INFO: ngram_model_dmp.c(649): 97 = #bigrams created
INFO: ngram_model_dmp.c(650): 3 = #prob2 entries
INFO: ngram_model_dmp.c(657): 3 = #bo_wt2 entries
INFO: ngram_model_dmp.c(661): 49 = #trigrams created
INFO: ngram_model_dmp.c(662): 2 = #prob3 entries
2015-11-01 00:42:22.352 OpenEarsSampleApp[2568:1401541] Done creating language model with CMUCLMTK in 0.077401 seconds.
INFO: cmd_ln.c(703): Parsing command line:
sphinx_lm_convert \
-i /var/mobile/Containers/Data/Application/6E796B17-868E-448C-8B0F-093B1FF028E8/Library/Caches/SecondOpenEarsDynamicLanguageModel.arpa \
-o /var/mobile/Containers/Data/Application/6E796B17-868E-448C-8B0F-093B1FF028E8/Library/Caches/SecondOpenEarsDynamicLanguageModel.DMPCurrent configuration:
[NAME] [DEFLT] [VALUE]
-case
-debug 0
-help no no
-i /var/mobile/Containers/Data/Application/6E796B17-868E-448C-8B0F-093B1FF028E8/Library/Caches/SecondOpenEarsDynamicLanguageModel.arpa
-ifmt
-logbase 1.0001 1.000100e+00
-mmap no no
-o /var/mobile/Containers/Data/Application/6E796B17-868E-448C-8B0F-093B1FF028E8/Library/Caches/SecondOpenEarsDynamicLanguageModel.DMP
-ofmtINFO: ngram_model_arpa.c(503): ngrams 1=51, 2=97, 3=49
INFO: ngram_model_arpa.c(135): Reading unigrams
INFO: ngram_model_arpa.c(542): 51 = #unigrams created
INFO: ngram_model_arpa.c(195): Reading bigrams
INFO: ngram_model_arpa.c(560): 97 = #bigrams created
INFO: ngram_model_arpa.c(561): 3 = #prob2 entries
INFO: ngram_model_arpa.c(569): 3 = #bo_wt2 entries
INFO: ngram_model_arpa.c(292): Reading trigrams
INFO: ngram_model_arpa.c(582): 49 = #trigrams created
INFO: ngram_model_arpa.c(583): 2 = #prob3 entries
INFO: ngram_model_dmp.c(518): Building DMP model…
INFO: ngram_model_dmp.c(548): 51 = #unigrams created
INFO: ngram_model_dmp.c(649): 97 = #bigrams created
INFO: ngram_model_dmp.c(650): 3 = #prob2 entries
INFO: ngram_model_dmp.c(657): 3 = #bo_wt2 entries
INFO: ngram_model_dmp.c(661): 49 = #trigrams created
INFO: ngram_model_dmp.c(662): 2 = #prob3 entries
2015-11-01 00:42:22.359 OpenEarsSampleApp[2568:1401541] I’m done running dynamic language model generation and it took 0.165418 seconds
2015-11-01 00:42:22.360 OpenEarsSampleApp[2568:1401541]Welcome to the OpenEars sample project. This project understands the words:
BACKWARD,
CHANGE,
FORWARD,
GO,
LEFT,
MODEL,
RIGHT,
TURN,
and if you say “CHANGE MODEL” it will switch to its dynamically-generated model which understands the words:
CHANGE,
MODEL,
MONDAY,
TUESDAY,
WEDNESDAY,
THURSDAY,
FRIDAY,
SATURDAY,
SUNDAY,
QUIDNUNC
2015-11-01 00:42:22.360 OpenEarsSampleApp[2568:1401541] Attempting to start listening session from startListeningWithLanguageModelAtPath:
2015-11-01 00:42:22.365 OpenEarsSampleApp[2568:1401541] User gave mic permission for this app.
2015-11-01 00:42:22.366 OpenEarsSampleApp[2568:1401541] setSecondsOfSilence wasn’t set, using default of 0.700000.
2015-11-01 00:42:22.367 OpenEarsSampleApp[2568:1401562] Starting listening.
2015-11-01 00:42:22.367 OpenEarsSampleApp[2568:1401562] about to set up audio session
2015-11-01 00:42:22.366 OpenEarsSampleApp[2568:1401541] Successfully started listening session from startListeningWithLanguageModelAtPath:
2015-11-01 00:42:22.368 OpenEarsSampleApp[2568:1401562] Creating audio session with default settings.
2015-11-01 00:42:22.408 OpenEarsSampleApp[2568:1401564] Audio route has changed for the following reason:
2015-11-01 00:42:22.411 OpenEarsSampleApp[2568:1401564] There was a category change. The new category is AVAudioSessionCategoryPlayAndRecord
2015-11-01 00:42:22.422 OpenEarsSampleApp[2568:1401564] This is not a case in which OpenEars notifies of a route change. At the close of this function, the new audio route is —SpeakerMicrophoneBuiltIn—. The previous route before changing to this route was <AVAudioSessionRouteDescription: 0x15dd8b0c0,
inputs = (null);
outputs = (
“<AVAudioSessionPortDescription: 0x15dd8b1e0, type = Speaker; name = Altavoz; UID = Speaker; selectedDataSource = (null)>”
)>.
2015-11-01 00:42:22.606 OpenEarsSampleApp[2568:1401562] done starting audio unit
INFO: cmd_ln.c(703): Parsing command line:
\
-lm /var/mobile/Containers/Data/Application/6E796B17-868E-448C-8B0F-093B1FF028E8/Library/Caches/FirstOpenEarsDynamicLanguageModel.DMP \
-vad_prespeech 10 \
-vad_postspeech 69 \
-vad_threshold 4.300000 \
-remove_noise yes \
-remove_silence yes \
-bestpath yes \
-lw 6.500000 \
-dict /var/mobile/Containers/Data/Application/6E796B17-868E-448C-8B0F-093B1FF028E8/Library/Caches/FirstOpenEarsDynamicLanguageModel.dic \
-hmm /var/mobile/Containers/Bundle/Application/CDC70BDD-C0A4-42C6-9B69-6E89E5B4B42F/OpenEarsSampleApp.app/AcousticModelSpanish.bundleCurrent configuration:
[NAME] [DEFLT] [VALUE]
-agc none none
-agcthresh 2.0 2.000000e+00
-allphone
-allphone_ci no no
-alpha 0.97 9.700000e-01
-argfile
-ascale 20.0 2.000000e+01
-aw 1 1
-backtrace no no
-beam 1e-48 1.000000e-48
-bestpath yes yes
-bestpathlw 9.5 9.500000e+00-ceplen 13 13
-cmn current current
-cmninit 8.0 8.0
-compallsen no no
-debug 0
-dict /var/mobile/Containers/Data/Application/6E796B17-868E-448C-8B0F-093B1FF028E8/Library/Caches/FirstOpenEarsDynamicLanguageModel.dic
-dictcase no no
-dither no no
-doublebw no no
-ds 1 1
-fdict
-feat 1s_c_d_dd 1s_c_d_dd
-featparams
-fillprob 1e-8 1.000000e-08
-frate 100 100
-fsg
-fsgusealtpron yes yes
-fsgusefiller yes yes
-fwdflat yes yes
-fwdflatbeam 1e-64 1.000000e-64
-fwdflatefwid 4 4
-fwdflatlw 8.5 8.500000e+00
-fwdflatsfwin 25 25
-fwdflatwbeam 7e-29 7.000000e-29
-fwdtree yes yes
-hmm /var/mobile/Containers/Bundle/Application/CDC70BDD-C0A4-42C6-9B69-6E89E5B4B42F/OpenEarsSampleApp.app/AcousticModelSpanish.bundle
-input_endian little little
-jsgf
-keyphrase
-kws
-kws_delay 10 10
-kws_plp 1e-1 1.000000e-01
-kws_threshold 1 1.000000e+00
-latsize 5000 5000
-lda
-ldadim 0 0
-lifter 0 0
-lm /var/mobile/Containers/Data/Application/6E796B17-868E-448C-8B0F-093B1FF028E8/Library/Caches/FirstOpenEarsDynamicLanguageModel.DMP
-lmctl
-lmname
-logbase 1.0001 1.000100e+00
-logfn
-logspec no no
-lowerf 133.33334 1.333333e+02
-lpbeam 1e-40 1.000000e-40
-lponlybeam 7e-29 7.000000e-29
-lw 6.5 6.500000e+00
-maxhmmpf 30000 30000
-maxwpf -1 -1
-mdef
-mean
-mfclogdir
-min_endfr 0 0
-mixw
-mixwfloor 0.0000001 1.000000e-07
-mllr
-mmap yes yes
-ncep 13 13
-nfft 512 512
-nfilt 40 40
-nwpen 1.0 1.000000e+00
-pbeam 1e-48 1.000000e-48
-pip 1.0 1.000000e+00
-pl_beam 1e-10 1.000000e-10
-pl_pbeam 1e-10 1.000000e-10
-pl_pip 1.0 1.000000e+00
-pl_weight 3.0 3.000000e+00
-pl_window 5 5
-rawlogdir
-remove_dc no no
-remove_noise yes yes
-remove_silence yes yes
-round_filters yes yes
-samprate 16000 1.600000e+04
-seed -1 -1
-sendump
-senlogdir
-senmgau
-silprob 0.005 5.000000e-03
-smoothspec no no
-svspec
-tmat
-tmatfloor 0.0001 1.000000e-04
-topn 4 4
-topn_beam 0 0
-toprule
-transform legacy legacy
-unit_area yes yes
-upperf 6855.4976 6.855498e+03
-uw 1.0 1.000000e+00
-vad_postspeech 50 69
-vad_prespeech 20 10
-vad_startspeech 10 10
-vad_threshold 2.0 4.300000e+00
-var
-varfloor 0.0001 1.000000e-04
-varnorm no no
-verbose no no
-warp_params
-warp_type inverse_linear inverse_linear
-wbeam 7e-29 7.000000e-29
-wip 0.65 6.500000e-01
-wlen 0.025625 2.562500e-02INFO: cmd_ln.c(703): Parsing command line:
\
-feat s3_1x39Current configuration:
[NAME] [DEFLT] [VALUE]
-agc none none
-agcthresh 2.0 2.000000e+00
-alpha 0.97 9.700000e-01
-ceplen 13 13
-cmn current current
-cmninit 8.0 8.0
-dither no no
-doublebw no no
-feat 1s_c_d_dd s3_1x39
-frate 100 100
-input_endian little little
-lda
-ldadim 0 0
-lifter 0 0
-logspec no no
-lowerf 133.33334 1.333333e+02
-ncep 13 13
-nfft 512 512
-nfilt 40 40
-remove_dc no no
-remove_noise yes yes
-remove_silence yes yes
-round_filters yes yes
-samprate 16000 1.600000e+04
-seed -1 -1
-smoothspec no no
-svspec
-transform legacy legacy
-unit_area yes yes
-upperf 6855.4976 6.855498e+03
-vad_postspeech 50 69
-vad_prespeech 20 10
-vad_startspeech 10 10
-vad_threshold 2.0 4.300000e+00
-varnorm no no
-verbose no no
-warp_params
-warp_type inverse_linear inverse_linear
-wlen 0.025625 2.562500e-02INFO: acmod.c(252): Parsed model-specific feature parameters from /var/mobile/Containers/Bundle/Application/CDC70BDD-C0A4-42C6-9B69-6E89E5B4B42F/OpenEarsSampleApp.app/AcousticModelSpanish.bundle/feat.params
INFO: feat.c(715): Initializing feature stream to type: ‘s3_1x39′, ceplen=13, CMN=’current’, VARNORM=’no’, AGC=’none’
INFO: cmn.c(143): mean[0]= 12.00, mean[1..12]= 0.0
INFO: mdef.c(518): Reading model definition: /var/mobile/Containers/Bundle/Application/CDC70BDD-C0A4-42C6-9B69-6E89E5B4B42F/OpenEarsSampleApp.app/AcousticModelSpanish.bundle/mdef
INFO: bin_mdef.c(181): Allocating 27954 * 8 bytes (218 KiB) for CD tree
INFO: tmat.c(206): Reading HMM transition probability matrices: /var/mobile/Containers/Bundle/Application/CDC70BDD-C0A4-42C6-9B69-6E89E5B4B42F/OpenEarsSampleApp.app/AcousticModelSpanish.bundle/transition_matrices
INFO: acmod.c(124): Attempting to use PTM computation module
INFO: ms_gauden.c(198): Reading mixture gaussian parameter: /var/mobile/Containers/Bundle/Application/CDC70BDD-C0A4-42C6-9B69-6E89E5B4B42F/OpenEarsSampleApp.app/AcousticModelSpanish.bundle/means
INFO: ms_gauden.c(292): 2630 codebook, 1 feature, size:
INFO: ms_gauden.c(294): 16×39
INFO: ms_gauden.c(198): Reading mixture gaussian parameter: /var/mobile/Containers/Bundle/Application/CDC70BDD-C0A4-42C6-9B69-6E89E5B4B42F/OpenEarsSampleApp.app/AcousticModelSpanish.bundle/variances
INFO: ms_gauden.c(292): 2630 codebook, 1 feature, size:
INFO: ms_gauden.c(294): 16×39
INFO: ms_gauden.c(354): 16 variance values floored
INFO: ptm_mgau.c(801): Number of codebooks exceeds 256: 2630
INFO: acmod.c(126): Attempting to use semi-continuous computation module
INFO: ms_gauden.c(198): Reading mixture gaussian parameter: /var/mobile/Containers/Bundle/Application/CDC70BDD-C0A4-42C6-9B69-6E89E5B4B42F/OpenEarsSampleApp.app/AcousticModelSpanish.bundle/means
INFO: ms_gauden.c(292): 2630 codebook, 1 feature, size:
INFO: ms_gauden.c(294): 16×39
INFO: ms_gauden.c(198): Reading mixture gaussian parameter: /var/mobile/Containers/Bundle/Application/CDC70BDD-C0A4-42C6-9B69-6E89E5B4B42F/OpenEarsSampleApp.app/AcousticModelSpanish.bundle/variances
INFO: ms_gauden.c(292): 2630 codebook, 1 feature, size:
INFO: ms_gauden.c(294): 16×39
INFO: ms_gauden.c(354): 16 variance values floored
INFO: acmod.c(128): Falling back to general multi-stream GMM computation
INFO: ms_gauden.c(198): Reading mixture gaussian parameter: /var/mobile/Containers/Bundle/Application/CDC70BDD-C0A4-42C6-9B69-6E89E5B4B42F/OpenEarsSampleApp.app/AcousticModelSpanish.bundle/means
INFO: ms_gauden.c(292): 2630 codebook, 1 feature, size:
INFO: ms_gauden.c(294): 16×39
INFO: ms_gauden.c(198): Reading mixture gaussian parameter: /var/mobile/Containers/Bundle/Application/CDC70BDD-C0A4-42C6-9B69-6E89E5B4B42F/OpenEarsSampleApp.app/AcousticModelSpanish.bundle/variances
INFO: ms_gauden.c(292): 2630 codebook, 1 feature, size:
INFO: ms_gauden.c(294): 16×39
INFO: ms_gauden.c(354): 16 variance values floored
INFO: ms_senone.c(149): Reading senone mixture weights: /var/mobile/Containers/Bundle/Application/CDC70BDD-C0A4-42C6-9B69-6E89E5B4B42F/OpenEarsSampleApp.app/AcousticModelSpanish.bundle/mixture_weights
INFO: ms_senone.c(200): Truncating senone logs3(pdf) values by 10 bits
INFO: ms_senone.c(207): Not transposing mixture weights in memory
INFO: ms_senone.c(268): Read mixture weights for 2630 senones: 1 features x 16 codewords
INFO: ms_senone.c(320): Mapping senones to individual codebooks
INFO: ms_mgau.c(141): The value of topn: 4
INFO: phone_loop_search.c(114): State beam -225 Phone exit beam -225 Insertion penalty 0
INFO: dict.c(320): Allocating 4127 * 32 bytes (128 KiB) for word entries
INFO: dict.c(333): Reading main dictionary: /var/mobile/Containers/Data/Application/6E796B17-868E-448C-8B0F-093B1FF028E8/Library/Caches/FirstOpenEarsDynamicLanguageModel.dic
INFO: dict.c(213): Allocated 0 KiB for strings, 0 KiB for phones
INFO: dict.c(336): 31 words read
INFO: dict2pid.c(396): Building PID tables for dictionary
INFO: dict2pid.c(406): Allocating 26^3 * 2 bytes (34 KiB) for word-initial triphones
INFO: dict2pid.c(132): Allocated 16432 bytes (16 KiB) for word-final triphones
INFO: dict2pid.c(196): Allocated 16432 bytes (16 KiB) for single-phone word triphones
INFO: ngram_model_arpa.c(77): No \data\ mark in LM file
INFO: ngram_model_dmp.c(166): Will use memory-mapped I/O for LM file
INFO: ngram_model_dmp.c(220): ngrams 1=33, 2=62, 3=31
INFO: ngram_model_dmp.c(266): 33 = LM.unigrams(+trailer) read
INFO: ngram_model_dmp.c(312): 62 = LM.bigrams(+trailer) read
INFO: ngram_model_dmp.c(338): 31 = LM.trigrams read
INFO: ngram_model_dmp.c(363): 3 = LM.prob2 entries read
INFO: ngram_model_dmp.c(383): 3 = LM.bo_wt2 entries read
INFO: ngram_model_dmp.c(403): 2 = LM.prob3 entries read
INFO: ngram_model_dmp.c(431): 1 = LM.tseg_base entries read
INFO: ngram_model_dmp.c(487): 33 = ascii word strings read
INFO: ngram_search_fwdtree.c(99): 4 unique initial diphones
INFO: ngram_search_fwdtree.c(148): 0 root, 0 non-root channels, 29 single-phone words
INFO: ngram_search_fwdtree.c(186): Creating search tree
INFO: ngram_search_fwdtree.c(192): before: 0 root, 0 non-root channels, 29 single-phone words
INFO: ngram_search_fwdtree.c(326): after: max nonroot chan increased to 147
INFO: ngram_search_fwdtree.c(339): after: 4 root, 19 non-root channels, 28 single-phone words
INFO: ngram_search_fwdflat.c(157): fwdflat: min_ef_width = 4, max_sf_win = 25
2015-11-01 00:42:23.051 OpenEarsSampleApp[2568:1401562] Listening.
2015-11-01 00:42:23.053 OpenEarsSampleApp[2568:1401562] Project has these words or phrases in its dictionary:
___REJ_Y
___REJ_X
___REJ_V
___REJ_U
___REJ_T
___REJ_S
___REJ_RR
___REJ_R
___REJ_P
___REJ_O
___REJ_N
___REJ_M
___REJ_LL
___REJ_L
___REJ_K
___REJ_J
___REJ_I
___REJ_GN
___REJ_G
___REJ_F
___REJ_E
___REJ_D
___REJ_CH
___REJ_B
___REJ_A
CABEZA
CAMBIAR
CASA
DERECHA
HOLA
SOBRE
2015-11-01 00:42:23.053 OpenEarsSampleApp[2568:1401562] Recognition loop has started
2015-11-01 00:42:23.077 OpenEarsSampleApp[2568:1401541] Local callback: Pocketsphinx is now listening.
2015-11-01 00:42:23.078 OpenEarsSampleApp[2568:1401541] Local callback: Pocketsphinx started.
2015-11-01 00:42:23.702 OpenEarsSampleApp[2568:1401561] Speech detected…
2015-11-01 00:42:23.703 OpenEarsSampleApp[2568:1401541] Local callback: Pocketsphinx has detected speech.
2015-11-01 00:42:26.294 OpenEarsSampleApp[2568:1401561] End of speech detected…
2015-11-01 00:42:26.296 OpenEarsSampleApp[2568:1401541] Local callback: Pocketsphinx has detected a second of silence, concluding an utterance.
INFO: cmn_prior.c(131): cmn_prior_update: from < 8.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 >
INFO: cmn_prior.c(149): cmn_prior_update: to < 9.40 0.05 -0.14 0.16 -0.17 -0.30 -0.23 -0.14 -0.06 -0.16 -0.06 -0.07 -0.09 >
INFO: ngram_search_fwdtree.c(1553): 3179 words recognized (12/fr)
INFO: ngram_search_fwdtree.c(1555): 51120 senones evaluated (198/fr)
INFO: ngram_search_fwdtree.c(1559): 11650 channels searched (45/fr), 995 1st, 8055 last
INFO: ngram_search_fwdtree.c(1562): 6433 words for which last channels evaluated (24/fr)
INFO: ngram_search_fwdtree.c(1564): 187 candidate words for entering last phone (0/fr)
INFO: ngram_search_fwdtree.c(1567): fwdtree 0.81 CPU 0.312 xRT
INFO: ngram_search_fwdtree.c(1570): fwdtree 3.24 wall 1.255 xRT
INFO: ngram_search_fwdflat.c(302): Utterance vocabulary contains 28 words
INFO: ngram_search_fwdflat.c(948): 2388 words recognized (9/fr)
INFO: ngram_search_fwdflat.c(950): 29634 senones evaluated (115/fr)
INFO: ngram_search_fwdflat.c(952): 7679 channels searched (29/fr)
INFO: ngram_search_fwdflat.c(954): 4569 words searched (17/fr)
INFO: ngram_search_fwdflat.c(957): 3362 word transitions (13/fr)
INFO: ngram_search_fwdflat.c(960): fwdflat 0.12 CPU 0.046 xRT
INFO: ngram_search_fwdflat.c(963): fwdflat 0.12 wall 0.047 xRT
INFO: ngram_search.c(1280): lattice start node <s>.0 end node </s>.220
INFO: ngram_search.c(1306): Eliminated 0 nodes before end node
INFO: ngram_search.c(1411): Lattice has 852 nodes, 9003 links
INFO: ps_lattice.c(1380): Bestpath score: -9394
INFO: ps_lattice.c(1384): Normalizer P(O) = alpha(</s>:220:256) = -594304
INFO: ps_lattice.c(1441): Joint P(O,S) = -609793 P(S|O) = -15489
INFO: ngram_search.c(899): bestpath 0.02 CPU 0.009 xRT
INFO: ngram_search.c(902): bestpath 0.03 wall 0.010 xRT
2015-11-01 00:42:26.456 OpenEarsSampleApp[2568:1401561] Pocketsphinx heard “HOLA” with a score of (-15489) and an utterance ID of 0.
2015-11-01 00:42:26.456 OpenEarsSampleApp[2568:1401541] Flite sending interrupt speech request.
2015-11-01 00:42:26.456 OpenEarsSampleApp[2568:1401541] Local callback: The received hypothesis is HOLA with a score of -15489 and an ID of 0
2015-11-01 00:42:26.458 OpenEarsSampleApp[2568:1401541] I’m running flite
2015-11-01 00:42:26.466 OpenEarsSampleApp[2568:1401561] Speech detected…
2015-11-01 00:42:26.508 OpenEarsSampleApp[2568:1401541] I’m done running flite and it took 0.050184 seconds
2015-11-01 00:42:26.508 OpenEarsSampleApp[2568:1401541] Flite audio player was nil when referenced so attempting to allocate a new audio player.
2015-11-01 00:42:26.509 OpenEarsSampleApp[2568:1401541] Loading speech data for Flite concluded successfully.
2015-11-01 00:42:26.591 OpenEarsSampleApp[2568:1401541] Local callback: Pocketsphinx has detected speech.
2015-11-01 00:42:26.592 OpenEarsSampleApp[2568:1401541] Flite sending suspend recognition notification.
2015-11-01 00:42:26.593 OpenEarsSampleApp[2568:1401541] Local callback: Flite has started speaking
2015-11-01 00:42:26.598 OpenEarsSampleApp[2568:1401541] Local callback: Pocketsphinx has suspended recognition.
2015-11-01 00:42:27.897 OpenEarsSampleApp[2568:1401541] AVAudioPlayer did finish playing with success flag of 1
2015-11-01 00:42:28.049 OpenEarsSampleApp[2568:1401541] Flite sending resume recognition notification.
2015-11-01 00:42:28.550 OpenEarsSampleApp[2568:1401541] Local callback: Flite has finished speaking
2015-11-01 00:42:28.558 OpenEarsSampleApp[2568:1401541] setSecondsOfSilence wasn’t set, using default of 0.700000.
2015-11-01 00:42:28.559 OpenEarsSampleApp[2568:1401541] Local callback: Pocketsphinx has resumed recognition.
INFO: cmn_prior.c(131): cmn_prior_update: from < 9.40 0.05 -0.14 0.16 -0.17 -0.30 -0.23 -0.14 -0.06 -0.16 -0.06 -0.07 -0.09 >
INFO: cmn_prior.c(149): cmn_prior_update: to < 9.78 0.13 -0.14 0.12 -0.24 -0.33 -0.22 -0.11 -0.06 -0.18 -0.07 -0.05 -0.09 >
INFO: ngram_search_fwdtree.c(1553): 382 words recognized (10/fr)
INFO: ngram_search_fwdtree.c(1555): 6979 senones evaluated (189/fr)
INFO: ngram_search_fwdtree.c(1559): 2032 channels searched (54/fr), 131 1st, 1621 last
INFO: ngram_search_fwdtree.c(1562): 701 words for which last channels evaluated (18/fr)
INFO: ngram_search_fwdtree.c(1564): 36 candidate words for entering last phone (0/fr)
INFO: ngram_search_fwdtree.c(1567): fwdtree 0.22 CPU 0.585 xRT
INFO: ngram_search_fwdtree.c(1570): fwdtree 2.27 wall 6.122 xRT
INFO: ngram_search_fwdflat.c(302): Utterance vocabulary contains 16 words
INFO: ngram_search_fwdflat.c(948): 189 words recognized (5/fr)
INFO: ngram_search_fwdflat.c(950): 5279 senones evaluated (143/fr)
INFO: ngram_search_fwdflat.c(952): 1739 channels searched (47/fr)
INFO: ngram_search_fwdflat.c(954): 597 words searched (16/fr)
INFO: ngram_search_fwdflat.c(957): 523 word transitions (14/fr)
INFO: ngram_search_fwdflat.c(960): fwdflat 0.03 CPU 0.079 xRT
INFO: ngram_search_fwdflat.c(963): fwdflat 0.03 wall 0.091 xRT
2015-11-01 00:42:29.613 OpenEarsSampleApp[2568:1401561] Speech detected…
2015-11-01 00:42:29.613 OpenEarsSampleApp[2568:1401541] Local callback: Pocketsphinx has detected speech.
2015-11-01 00:42:30.356 OpenEarsSampleApp[2568:1401561] End of speech detected…
2015-11-01 00:42:30.356 OpenEarsSampleApp[2568:1401541] Local callback: Pocketsphinx has detected a second of silence, concluding an utterance.
INFO: cmn_prior.c(131): cmn_prior_update: from < 9.78 0.13 -0.14 0.12 -0.24 -0.33 -0.22 -0.11 -0.06 -0.18 -0.07 -0.05 -0.09 >
INFO: cmn_prior.c(149): cmn_prior_update: to < 9.66 0.22 -0.11 0.07 -0.27 -0.34 -0.18 -0.10 -0.05 -0.15 -0.07 -0.05 -0.08 >
INFO: ngram_search_fwdtree.c(1553): 1471 words recognized (17/fr)
INFO: ngram_search_fwdtree.c(1555): 16769 senones evaluated (188/fr)
INFO: ngram_search_fwdtree.c(1559): 3783 channels searched (42/fr), 340 1st, 2770 last
INFO: ngram_search_fwdtree.c(1562): 2288 words for which last channels evaluated (25/fr)
INFO: ngram_search_fwdtree.c(1564): 28 candidate words for entering last phone (0/fr)
INFO: ngram_search_fwdtree.c(1567): fwdtree 0.24 CPU 0.275 xRT
INFO: ngram_search_fwdtree.c(1570): fwdtree 1.61 wall 1.810 xRT
INFO: ngram_search_fwdflat.c(302): Utterance vocabulary contains 21 words
INFO: ngram_search_fwdflat.c(948): 903 words recognized (10/fr)
INFO: ngram_search_fwdflat.c(950): 9712 senones evaluated (109/fr)
INFO: ngram_search_fwdflat.c(952): 2525 channels searched (28/fr)
INFO: ngram_search_fwdflat.c(954): 1502 words searched (16/fr)
INFO: ngram_search_fwdflat.c(957): 1191 word transitions (13/fr)
INFO: ngram_search_fwdflat.c(960): fwdflat 0.05 CPU 0.056 xRT
INFO: ngram_search_fwdflat.c(963): fwdflat 0.05 wall 0.057 xRT
INFO: ngram_search.c(1280): lattice start node <s>.0 end node </s>.74
INFO: ngram_search.c(1306): Eliminated 0 nodes before end node
INFO: ngram_search.c(1411): Lattice has 315 nodes, 2137 links
INFO: ps_lattice.c(1380): Bestpath score: -2356
INFO: ps_lattice.c(1384): Normalizer P(O) = alpha(</s>:74:87) = -170735
INFO: ps_lattice.c(1441): Joint P(O,S) = -182742 P(S|O) = -12007
INFO: ngram_search.c(899): bestpath 0.01 CPU 0.006 xRT
INFO: ngram_search.c(902): bestpath 0.00 wall 0.005 xRT
2015-11-01 00:42:30.424 OpenEarsSampleApp[2568:1401561] Pocketsphinx heard ” ” with a score of (-12007) and an utterance ID of 1.
2015-11-01 00:42:30.424 OpenEarsSampleApp[2568:1401561] Hypothesis was null so we aren’t returning it. If you want null hypotheses to also be returned, set OEPocketsphinxController’s property returnNullHypotheses to TRUE before starting OEPocketsphinxController.
2015-11-01 00:42:30.506 OpenEarsSampleApp[2568:1401561] Speech detected…
2015-11-01 00:42:30.507 OpenEarsSampleApp[2568:1401541] Local callback: Pocketsphinx has detected speech.
2015-11-01 00:42:31.536 OpenEarsSampleApp[2568:1401561] End of speech detected…
2015-11-01 00:42:31.536 OpenEarsSampleApp[2568:1401541] Local callback: Pocketsphinx has detected a second of silence, concluding an utterance.
INFO: cmn_prior.c(131): cmn_prior_update: from < 9.66 0.22 -0.11 0.07 -0.27 -0.34 -0.18 -0.10 -0.05 -0.15 -0.07 -0.05 -0.08 >
INFO: cmn_prior.c(149): cmn_prior_update: to < 9.68 0.28 -0.14 0.03 -0.24 -0.35 -0.20 -0.11 -0.06 -0.14 -0.06 -0.05 -0.09 >
INFO: ngram_search_fwdtree.c(1553): 1664 words recognized (16/fr)
INFO: ngram_search_fwdtree.c(1555): 18900 senones evaluated (180/fr)
INFO: ngram_search_fwdtree.c(1559): 4315 channels searched (41/fr), 371 1st, 3141 last
INFO: ngram_search_fwdtree.c(1562): 2517 words for which last channels evaluated (23/fr)
INFO: ngram_search_fwdtree.c(1564): 58 candidate words for entering last phone (0/fr)
INFO: ngram_search_fwdtree.c(1567): fwdtree 0.28 CPU 0.268 xRT
INFO: ngram_search_fwdtree.c(1570): fwdtree 1.12 wall 1.070 xRT
INFO: ngram_search_fwdflat.c(302): Utterance vocabulary contains 24 words
INFO: ngram_search_fwdflat.c(948): 1010 words recognized (10/fr)
INFO: ngram_search_fwdflat.c(950): 12662 senones evaluated (121/fr)
INFO: ngram_search_fwdflat.c(952): 3285 channels searched (31/fr)
INFO: ngram_search_fwdflat.c(954): 1944 words searched (18/fr)
INFO: ngram_search_fwdflat.c(957): 1532 word transitions (14/fr)
INFO: ngram_search_fwdflat.c(960): fwdflat 0.05 CPU 0.051 xRT
INFO: ngram_search_fwdflat.c(963): fwdflat 0.06 wall 0.053 xRT
INFO: ngram_search.c(1280): lattice start node <s>.0 end node </s>.90
INFO: ngram_search.c(1306): Eliminated 0 nodes before end node
INFO: ngram_search.c(1411): Lattice has 360 nodes, 2459 links
INFO: ps_lattice.c(1380): Bestpath score: -3860
INFO: ps_lattice.c(1384): Normalizer P(O) = alpha(</s>:90:103) = -261477
INFO: ps_lattice.c(1441): Joint P(O,S) = -295221 P(S|O) = -33744
INFO: ngram_search.c(899): bestpath 0.01 CPU 0.009 xRT
INFO: ngram_search.c(902): bestpath 0.01 wall 0.006 xRT
2015-11-01 00:42:31.611 OpenEarsSampleApp[2568:1401561] Pocketsphinx heard “HOLA” with a score of (-33744) and an utterance ID of 2.
2015-11-01 00:42:31.611 OpenEarsSampleApp[2568:1401541] Flite sending interrupt speech request.
2015-11-01 00:42:31.611 OpenEarsSampleApp[2568:1401541] Local callback: The received hypothesis is HOLA with a score of -33744 and an ID of 2
2015-11-01 00:42:31.611 OpenEarsSampleApp[2568:1401541] I’m running flite
2015-11-01 00:42:31.662 OpenEarsSampleApp[2568:1401541] I’m done running flite and it took 0.050604 seconds
2015-11-01 00:42:31.662 OpenEarsSampleApp[2568:1401541] Flite audio player was nil when referenced so attempting to allocate a new audio player.
2015-11-01 00:42:31.662 OpenEarsSampleApp[2568:1401541] Loading speech data for Flite concluded successfully.
2015-11-01 00:42:31.699 OpenEarsSampleApp[2568:1401541] Flite sending suspend recognition notification.
2015-11-01 00:42:31.700 OpenEarsSampleApp[2568:1401541] Local callback: Flite has started speaking
2015-11-01 00:42:31.704 OpenEarsSampleApp[2568:1401541] Local callback: Pocketsphinx has suspended recognition.
2015-11-01 00:42:32.913 OpenEarsSampleApp[2568:1401541] AVAudioPlayer did finish playing with success flag of 1
2015-11-01 00:42:33.065 OpenEarsSampleApp[2568:1401541] Flite sending resume recognition notification.
2015-11-01 00:42:33.566 OpenEarsSampleApp[2568:1401541] Local callback: Flite has finished speaking
2015-11-01 00:42:33.574 OpenEarsSampleApp[2568:1401541] setSecondsOfSilence wasn’t set, using default of 0.700000.
2015-11-01 00:42:33.575 OpenEarsSampleApp[2568:1401541] Local callback: Pocketsphinx has resumed recognition.
INFO: cmn_prior.c(131): cmn_prior_update: from < 9.68 0.28 -0.14 0.03 -0.24 -0.35 -0.20 -0.11 -0.06 -0.14 -0.06 -0.05 -0.09 >
INFO: cmn_prior.c(149): cmn_prior_update: to < 9.68 0.28 -0.14 0.03 -0.24 -0.35 -0.20 -0.11 -0.06 -0.14 -0.06 -0.05 -0.09 >
INFO: ngram_search_fwdflat.c(302): Utterance vocabulary contains 0 words
2015-11-01 00:42:35.483 OpenEarsSampleApp[2568:1401562] Speech detected…
2015-11-01 00:42:35.484 OpenEarsSampleApp[2568:1401541] Local callback: Pocketsphinx has detected speech.
2015-11-01 00:42:36.140 OpenEarsSampleApp[2568:1401562] End of speech detected…
INFO: cmn_prior.c(131): cmn_prior_update: from < 9.68 0.28 -0.14 0.03 -0.24 -0.35 -0.20 -0.11 -0.06 -0.14 -0.06 -0.05 -0.09 >
INFO: cmn_prior.c(149): cmn_prior_update: to < 9.66 0.29 -0.16 -0.01 -0.27 -0.32 -0.19 -0.11 -0.05 -0.15 -0.05 -0.06 -0.10 >
2015-11-01 00:42:36.141 OpenEarsSampleApp[2568:1401541] Local callback: Pocketsphinx has detected a second of silence, concluding an utterance.
INFO: ngram_search_fwdtree.c(1553): 1119 words recognized (14/fr)
INFO: ngram_search_fwdtree.c(1555): 14835 senones evaluated (188/fr)
INFO: ngram_search_fwdtree.c(1559): 4115 channels searched (52/fr), 263 1st, 3149 last
INFO: ngram_search_fwdtree.c(1562): 1819 words for which last channels evaluated (23/fr)
INFO: ngram_search_fwdtree.c(1564): 59 candidate words for entering last phone (0/fr)
INFO: ngram_search_fwdtree.c(1567): fwdtree 0.30 CPU 0.382 xRT
INFO: ngram_search_fwdtree.c(1570): fwdtree 2.46 wall 3.110 xRT
INFO: ngram_search_fwdflat.c(302): Utterance vocabulary contains 19 words
INFO: ngram_search_fwdflat.c(948): 596 words recognized (8/fr)
INFO: ngram_search_fwdflat.c(950): 10555 senones evaluated (134/fr)
INFO: ngram_search_fwdflat.c(952): 4103 channels searched (51/fr)
INFO: ngram_search_fwdflat.c(954): 1310 words searched (16/fr)
INFO: ngram_search_fwdflat.c(957): 949 word transitions (12/fr)
INFO: ngram_search_fwdflat.c(960): fwdflat 0.04 CPU 0.048 xRT
INFO: ngram_search_fwdflat.c(963): fwdflat 0.05 wall 0.059 xRT
INFO: ngram_search.c(1280): lattice start node <s>.0 end node </s>.68
INFO: ngram_search.c(1306): Eliminated 1 nodes before end node
INFO: ngram_search.c(1411): Lattice has 157 nodes, 520 links
INFO: ps_lattice.c(1380): Bestpath score: -2993
INFO: ps_lattice.c(1384): Normalizer P(O) = alpha(</s>:68:77) = -180236
INFO: ps_lattice.c(1441): Joint P(O,S) = -192013 P(S|O) = -11777
INFO: ngram_search.c(899): bestpath 0.00 CPU 0.002 xRT
INFO: ngram_search.c(902): bestpath 0.00 wall 0.002 xRT
2015-11-01 00:42:36.199 OpenEarsSampleApp[2568:1401562] Pocketsphinx heard “CABEZA” with a score of (-11777) and an utterance ID of 3.
2015-11-01 00:42:36.199 OpenEarsSampleApp[2568:1401541] Flite sending interrupt speech request.
2015-11-01 00:42:36.200 OpenEarsSampleApp[2568:1401541] Local callback: The received hypothesis is CABEZA with a score of -11777 and an ID of 3
2015-11-01 00:42:36.201 OpenEarsSampleApp[2568:1401541] I’m running flite
2015-11-01 00:42:36.263 OpenEarsSampleApp[2568:1401541] I’m done running flite and it took 0.062081 seconds
2015-11-01 00:42:36.263 OpenEarsSampleApp[2568:1401541] Flite audio player was nil when referenced so attempting to allocate a new audio player.
2015-11-01 00:42:36.263 OpenEarsSampleApp[2568:1401541] Loading speech data for Flite concluded successfully.
2015-11-01 00:42:36.345 OpenEarsSampleApp[2568:1401541] Flite sending suspend recognition notification.
2015-11-01 00:42:36.348 OpenEarsSampleApp[2568:1401541] Local callback: Flite has started speaking
2015-11-01 00:42:36.357 OpenEarsSampleApp[2568:1401541] Local callback: Pocketsphinx has suspended recognition.
2015-11-01 00:42:36.378 OpenEarsSampleApp[2568:1401562] Speech detected…
2015-11-01 00:42:36.379 OpenEarsSampleApp[2568:1401541] Local callback: Pocketsphinx has detected speech.
2015-11-01 00:42:37.648 OpenEarsSampleApp[2568:1401541] AVAudioPlayer did finish playing with success flag of 1
2015-11-01 00:42:37.800 OpenEarsSampleApp[2568:1401541] Flite sending resume recognition notification.
2015-11-01 00:42:38.302 OpenEarsSampleApp[2568:1401541] Local callback: Flite has finished speaking
2015-11-01 00:42:38.310 OpenEarsSampleApp[2568:1401541] setSecondsOfSilence wasn’t set, using default of 0.700000.
2015-11-01 00:42:38.310 OpenEarsSampleApp[2568:1401541] Local callback: Pocketsphinx has resumed recognition.
INFO: cmn_prior.c(131): cmn_prior_update: from < 9.66 0.29 -0.16 -0.01 -0.27 -0.32 -0.19 -0.11 -0.05 -0.15 -0.05 -0.06 -0.10 >
INFO: cmn_prior.c(149): cmn_prior_update: to < 9.77 0.31 -0.15 -0.01 -0.28 -0.32 -0.19 -0.12 -0.05 -0.15 -0.05 -0.06 -0.10 >
INFO: ngram_search_fwdtree.c(1553): 62 words recognized (4/fr)
INFO: ngram_search_fwdtree.c(1555): 1194 senones evaluated (75/fr)
INFO: ngram_search_fwdtree.c(1559): 290 channels searched (18/fr), 36 1st, 230 last
INFO: ngram_search_fwdtree.c(1562): 167 words for which last channels evaluated (10/fr)
INFO: ngram_search_fwdtree.c(1564): 4 candidate words for entering last phone (0/fr)
INFO: ngram_search_fwdtree.c(1567): fwdtree 0.19 CPU 1.207 xRT
INFO: ngram_search_fwdtree.c(1570): fwdtree 2.24 wall 13.999 xRT
INFO: ngram_search_fwdflat.c(302): Utterance vocabulary contains 6 words
INFO: ngram_search_fwdflat.c(948): 54 words recognized (3/fr)
INFO: ngram_search_fwdflat.c(950): 447 senones evaluated (28/fr)
INFO: ngram_search_fwdflat.c(952): 99 channels searched (6/fr)
INFO: ngram_search_fwdflat.c(954): 99 words searched (6/fr)
INFO: ngram_search_fwdflat.c(957): 90 word transitions (5/fr)
INFO: ngram_search_fwdflat.c(960): fwdflat 0.01 CPU 0.057 xRT
INFO: ngram_search_fwdflat.c(963): fwdflat 0.01 wall 0.049 xRT
2015-11-01 00:42:40.762 OpenEarsSampleApp[2568:1401561] Speech detected…
2015-11-01 00:42:40.762 OpenEarsSampleApp[2568:1401541] Local callback: Pocketsphinx has detected speech.
2015-11-01 00:42:42.024 OpenEarsSampleApp[2568:1401561] End of speech detected…
2015-11-01 00:42:42.024 OpenEarsSampleApp[2568:1401541] Local callback: Pocketsphinx has detected a second of silence, concluding an utterance.
INFO: cmn_prior.c(131): cmn_prior_update: from < 9.77 0.31 -0.15 -0.01 -0.28 -0.32 -0.19 -0.12 -0.05 -0.15 -0.05 -0.06 -0.10 >
INFO: cmn_prior.c(149): cmn_prior_update: to < 9.57 0.24 -0.11 0.04 -0.26 -0.29 -0.19 -0.13 -0.05 -0.16 -0.06 -0.07 -0.10 >
INFO: ngram_search_fwdtree.c(1553): 2365 words recognized (17/fr)
INFO: ngram_search_fwdtree.c(1555): 28522 senones evaluated (204/fr)
INFO: ngram_search_fwdtree.c(1559): 6639 channels searched (47/fr), 544 1st, 4857 last
INFO: ngram_search_fwdtree.c(1562): 3632 words for which last channels evaluated (25/fr)
INFO: ngram_search_fwdtree.c(1564): 74 candidate words for entering last phone (0/fr)
INFO: ngram_search_fwdtree.c(1567): fwdtree 0.54 CPU 0.386 xRT
INFO: ngram_search_fwdtree.c(1570): fwdtree 3.59 wall 2.562 xRT
INFO: ngram_search_fwdflat.c(302): Utterance vocabulary contains 30 words
INFO: ngram_search_fwdflat.c(948): 1767 words recognized (13/fr)
INFO: ngram_search_fwdflat.c(950): 25338 senones evaluated (181/fr)
INFO: ngram_search_fwdflat.c(952): 7207 channels searched (51/fr)
INFO: ngram_search_fwdflat.c(954): 3345 words searched (23/fr)
INFO: ngram_search_fwdflat.c(957): 2537 word transitions (18/fr)
INFO: ngram_search_fwdflat.c(960): fwdflat 0.10 CPU 0.072 xRT
INFO: ngram_search_fwdflat.c(963): fwdflat 0.10 wall 0.072 xRT
INFO: ngram_search.c(1280): lattice start node <s>.0 end node </s>.113
INFO: ngram_search.c(1306): Eliminated 0 nodes before end node
INFO: ngram_search.c(1411): Lattice has 563 nodes, 5380 links
INFO: ps_lattice.c(1380): Bestpath score: -5524
INFO: ps_lattice.c(1384): Normalizer P(O) = alpha(</s>:113:138) = -361639
INFO: ps_lattice.c(1441): Joint P(O,S) = -404835 P(S|O) = -43196
INFO: ngram_search.c(899): bestpath 0.01 CPU 0.011 xRT
INFO: ngram_search.c(902): bestpath 0.01 wall 0.010 xRT
2015-11-01 00:42:42.151 OpenEarsSampleApp[2568:1401561] Pocketsphinx heard ” ” with a score of (-43196) and an utterance ID of 4.
2015-11-01 00:42:42.151 OpenEarsSampleApp[2568:1401561] Hypothesis was null so we aren’t returning it. If you want null hypotheses to also be returned, set OEPocketsphinxController’s property returnNullHypotheses to TRUE before starting OEPocketsphinxController.
2015-11-01 00:42:43.053 OpenEarsSampleApp[2568:1401561] Speech detected…
2015-11-01 00:42:43.054 OpenEarsSampleApp[2568:1401541] Local callback: Pocketsphinx has detected speech.
INFO: cmn_prior.c(99): cmn_prior_update: from < 9.57 0.24 -0.11 0.04 -0.26 -0.29 -0.19 -0.13 -0.05 -0.16 -0.06 -0.07 -0.10 >
INFO: cmn_prior.c(116): cmn_prior_update: to < 9.54 0.22 -0.10 0.08 -0.25 -0.30 -0.20 -0.14 -0.06 -0.17 -0.07 -0.08 -0.10 >
2015-11-01 00:42:43.840 OpenEarsSampleApp[2568:1401561] End of speech detected…
INFO: cmn_prior.c(131): cmn_prior_update: from < 9.54 0.22 -0.10 0.08 -0.25 -0.30 -0.20 -0.14 -0.06 -0.17 -0.07 -0.08 -0.10 >
INFO: cmn_prior.c(149): cmn_prior_update: to < 9.23 0.22 -0.09 0.08 2015-11-01 00:42:43.841 OpenEarsSampleApp[2568:1401541] Local callback: Pocketsphinx has detected a second of silence, concluding an utterance.
-0.24 -0.29 -0.20 -0.14 -0.06 -0.16 -0.07 -0.08 -0.10 >
INFO: ngram_search_fwdtree.c(1553): 1406 words recognized (18/fr)
INFO: ngram_search_fwdtree.c(1555): 13702 senones evaluated (176/fr)
INFO: ngram_search_fwdtree.c(1559): 2957 channels searched (37/fr), 296 1st, 2023 last
INFO: ngram_search_fwdtree.c(1562): 1981 words for which last channels evaluated (25/fr)
INFO: ngram_search_fwdtree.c(1564): 4 candidate words for entering last phone (0/fr)
INFO: ngram_search_fwdtree.c(1567): fwdtree 0.29 CPU 0.369 xRT
INFO: ngram_search_fwdtree.c(1570): fwdtree 1.70 wall 2.180 xRT
INFO: ngram_search_fwdflat.c(302): Utterance vocabulary contains 20 words
INFO: ngram_search_fwdflat.c(948): 1004 words recognized (13/fr)
INFO: ngram_search_fwdflat.c(950): 6714 senones evaluated (86/fr)
INFO: ngram_search_fwdflat.c(952): 1458 channels searched (18/fr)
INFO: ngram_search_fwdflat.c(954): 1458 words searched (18/fr)
INFO: ngram_search_fwdflat.c(957): 1117 word transitions (14/fr)
INFO: ngram_search_fwdflat.c(960): fwdflat 0.02 CPU 0.030 xRT
INFO: ngram_search_fwdflat.c(963): fwdflat 0.03 wall 0.041 xRT
INFO: ngram_search.c(1280): lattice start node <s>.0 end node </s>.68
INFO: ngram_search.c(1306): Eliminated 0 nodes before end node
INFO: ngram_search.c(1411): Lattice has 293 nodes, 2639 links
INFO: ps_lattice.c(1380): Bestpath score: -2509
INFO: ps_lattice.c(1384): Normalizer P(O) = alpha(</s>:68:76) = -159881
INFO: ps_lattice.c(1441): Joint P(O,S) = -181003 P(S|O) = -21122
INFO: ngram_search.c(899): bestpath 0.01 CPU 0.010 xRT
INFO: ngram_search.c(902): bestpath 0.01 wall 0.009 xRT
2015-11-01 00:42:43.891 OpenEarsSampleApp[2568:1401561] Pocketsphinx heard ” ” with a score of (-21122) and an utterance ID of 5.
2015-11-01 00:42:43.892 OpenEarsSampleApp[2568:1401561] Hypothesis was null so we aren’t returning it. If you want null hypotheses to also be returned, set OEPocketsphinxController’s property returnNullHypotheses to TRUE before starting OEPocketsphinxController.
2015-11-01 00:42:44.076 OpenEarsSampleApp[2568:1401561] Speech detected…
2015-11-01 00:42:44.076 OpenEarsSampleApp[2568:1401541] Local callback: Pocketsphinx has detected speech.
2015-11-01 00:42:44.841 OpenEarsSampleApp[2568:1401561] End of speech detected…
2015-11-01 00:42:44.842 OpenEarsSampleApp[2568:1401541] Local callback: Pocketsphinx has detected a second of silence, concluding an utterance.
INFO: cmn_prior.c(131): cmn_prior_update: from < 9.23 0.22 -0.09 0.08 -0.24 -0.29 -0.20 -0.14 -0.06 -0.16 -0.07 -0.08 -0.10 >
INFO: cmn_prior.c(149): cmn_prior_update: to < 9.14 0.27 -0.05 0.07 -0.29 -0.31 -0.17 -0.14 -0.05 -0.16 -0.06 -0.08 -0.11 >
INFO: ngram_search_fwdtree.c(1553): 1213 words recognized (16/fr)
INFO: ngram_search_fwdtree.c(1555): 14089 senones evaluated (183/fr)
INFO: ngram_search_fwdtree.c(1559): 3143 channels searched (40/fr), 292 1st, 2277 last
INFO: ngram_search_fwdtree.c(1562): 1899 words for which last channels evaluated (24/fr)
INFO: ngram_search_fwdtree.c(1564): 21 candidate words for entering last phone (0/fr)
INFO: ngram_search_fwdtree.c(1567): fwdtree 0.24 CPU 0.312 xRT
INFO: ngram_search_fwdtree.c(1570): fwdtree 0.97 wall 1.258 xRT
INFO: ngram_search_fwdflat.c(302): Utterance vocabulary contains 21 words
INFO: ngram_search_fwdflat.c(948): 752 words recognized (10/fr)
INFO: ngram_search_fwdflat.c(950): 9032 senones evaluated (117/fr)
INFO: ngram_search_fwdflat.c(952): 2280 channels searched (29/fr)
INFO: ngram_search_fwdflat.c(954): 1435 words searched (18/fr)
INFO: ngram_search_fwdflat.c(957): 1081 word transitions (14/fr)
INFO: ngram_search_fwdflat.c(960): fwdflat 0.05 CPU 0.059 xRT
INFO: ngram_search_fwdflat.c(963): fwdflat 0.05 wall 0.070 xRT
INFO: ngram_search.c(1280): lattice start node <s>.0 end node </s>.59
INFO: ngram_search.c(1306): Eliminated 0 nodes before end node
INFO: ngram_search.c(1411): Lattice has 263 nodes, 1187 links
INFO: ps_lattice.c(1380): Bestpath score: -2763
INFO: ps_lattice.c(1384): Normalizer P(O) = alpha(</s>:59:75) = -193073
INFO: ps_lattice.c(1441): Joint P(O,S) = -226567 P(S|O) = -33494
INFO: ngram_search.c(899): bestpath 0.00 CPU 0.004 xRT
INFO: ngram_search.c(902): bestpath 0.00 wall 0.004 xRT
2015-11-01 00:42:44.919 OpenEarsSampleApp[2568:1401561] Pocketsphinx heard ” ” with a score of (-33494) and an utterance ID of 6.
2015-11-01 00:42:44.919 OpenEarsSampleApp[2568:1401561] Hypothesis was null so we aren’t returning it. If you want null hypotheses to also be returned, set OEPocketsphinxController’s property returnNullHypotheses to TRUE before starting OEPocketsphinxController.
2015-11-01 00:42:48.837 OpenEarsSampleApp[2568:1401562] Speech detected…
2015-11-01 00:42:48.838 OpenEarsSampleApp[2568:1401541] Local callback: Pocketsphinx has detected speech.
2015-11-01 00:42:49.786 OpenEarsSampleApp[2568:1401562] End of speech detected…
INFO: cmn_prior.c(131): cmn_prior_update: from < 9.14 0.27 -0.05 0.07 -0.29 -0.31 -0.17 -0.14 -0.05 -0.16 -0.06 -0.08 -0.11 >
INFO: cmn_prior.c(149): cmn_prior_update: to < 9.02 0.32 0.01 0.10 -0.29 -0.31 -0.17 -0.14 -0.05 -0.18 2015-11-01 00:42:49.787 OpenEarsSampleApp[2568:1401541] Local callback: Pocketsphinx has detected a second of silence, concluding an utterance.
-0.07 -0.08 -0.12 >
INFO: ngram_search_fwdtree.c(1553): 1445 words recognized (13/fr)
INFO: ngram_search_fwdtree.c(1555): 17183 senones evaluated (159/fr)
INFO: ngram_search_fwdtree.c(1559): 3826 channels searched (35/fr), 383 1st, 2861 last
INFO: ngram_search_fwdtree.c(1562): 2384 words for which last channels evaluated (22/fr)
INFO: ngram_search_fwdtree.c(1564): 15 candidate words for entering last phone (0/fr)
INFO: ngram_search_fwdtree.c(1567): fwdtree 0.49 CPU 0.450 xRT
INFO: ngram_search_fwdtree.c(1570): fwdtree 4.88 wall 4.517 xRT
INFO: ngram_search_fwdflat.c(302): Utterance vocabulary contains 26 words
INFO: ngram_search_fwdflat.c(948): 817 words recognized (8/fr)
INFO: ngram_search_fwdflat.c(950): 11740 senones evaluated (109/fr)
INFO: ngram_search_fwdflat.c(952): 3046 channels searched (28/fr)
INFO: ngram_search_fwdflat.c(954): 1785 words searched (16/fr)
INFO: ngram_search_fwdflat.c(957): 1467 word transitions (13/fr)
INFO: ngram_search_fwdflat.c(960): fwdflat 0.06 CPU 0.052 xRT
INFO: ngram_search_fwdflat.c(963): fwdflat 0.06 wall 0.051 xRT
INFO: ngram_search.c(1280): lattice start node <s>.0 end node </s>.67
INFO: ngram_search.c(1306): Eliminated 0 nodes before end node
INFO: ngram_search.c(1411): Lattice has 193 nodes, 436 links
INFO: ps_lattice.c(1380): Bestpath score: -3055
INFO: ps_lattice.c(1384): Normalizer P(O) = alpha(</s>:67:106) = -235063
INFO: ps_lattice.c(1441): Joint P(O,S) = -241366 P(S|O) = -6303
INFO: ngram_search.c(899): bestpath 0.00 CPU 0.001 xRT
INFO: ngram_search.c(902): bestpath 0.00 wall 0.002 xRT
2015-11-01 00:42:49.855 OpenEarsSampleApp[2568:1401562] Pocketsphinx heard “DERECHA” with a score of (-6303) and an utterance ID of 7.
2015-11-01 00:42:49.856 OpenEarsSampleApp[2568:1401541] Flite sending interrupt speech request.
2015-11-01 00:42:49.856 OpenEarsSampleApp[2568:1401541] Local callback: The received hypothesis is DERECHA with a score of -6303 and an ID of 7
2015-11-01 00:42:49.857 OpenEarsSampleApp[2568:1401541] I’m running flite
2015-11-01 00:42:49.911 OpenEarsSampleApp[2568:1401541] I’m done running flite and it took 0.053779 seconds
2015-11-01 00:42:49.911 OpenEarsSampleApp[2568:1401541] Flite audio player was nil when referenced so attempting to allocate a new audio player.
2015-11-01 00:42:49.911 OpenEarsSampleApp[2568:1401541] Loading speech data for Flite concluded successfully.
2015-11-01 00:42:49.950 OpenEarsSampleApp[2568:1401562] Speech detected…
2015-11-01 00:42:49.997 OpenEarsSampleApp[2568:1401541] Flite sending suspend recognition notification.
2015-11-01 00:42:49.997 OpenEarsSampleApp[2568:1401541] Local callback: Pocketsphinx has detected speech.
2015-11-01 00:42:49.999 OpenEarsSampleApp[2568:1401541] Local callback: Flite has started speaking
2015-11-01 00:42:50.003 OpenEarsSampleApp[2568:1401541] Local callback: Pocketsphinx has suspended recognition.
2015-11-01 00:42:51.209 OpenEarsSampleApp[2568:1401541] AVAudioPlayer did finish playing with success flag of 1
2015-11-01 00:42:51.361 OpenEarsSampleApp[2568:1401541] Flite sending resume recognition notification.
2015-11-01 00:42:51.863 OpenEarsSampleApp[2568:1401541] Local callback: Flite has finished speaking
2015-11-01 00:42:51.871 OpenEarsSampleApp[2568:1401541] setSecondsOfSilence wasn’t set, using default of 0.700000.
2015-11-01 00:42:51.871 OpenEarsSampleApp[2568:1401541] Local callback: Pocketsphinx has resumed recognition.
INFO: cmn_prior.c(131): cmn_prior_update: from < 9.02 0.32 0.01 0.10 -0.29 -0.31 -0.17 -0.14 -0.05 -0.18 -0.07 -0.08 -0.12 >
INFO: cmn_prior.c(149): cmn_prior_update: to < 9.16 0.33 0.01 0.11 -0.29 -0.31 -0.18 -0.15 -0.05 -0.18 -0.07 -0.09 -0.12 >
INFO: ngram_search_fwdtree.c(1553): 265 words recognized (11/fr)
INFO: ngram_search_fwdtree.c(1555): 3787 senones evaluated (151/fr)
INFO: ngram_search_fwdtree.c(1559): 844 channels searched (33/fr), 84 1st, 599 last
INFO: ngram_search_fwdtree.c(1562): 536 words for which last channels evaluated (21/fr)
INFO: ngram_search_fwdtree.c(1564): 6 candidate words for entering last phone (0/fr)
INFO: ngram_search_fwdtree.c(1567): fwdtree 0.19 CPU 0.761 xRT
INFO: ngram_search_fwdtree.c(1570): fwdtree 2.17 wall 8.666 xRT
INFO: ngram_search_fwdflat.c(302): Utterance vocabulary contains 10 words
INFO: ngram_search_fwdflat.c(948): 157 words recognized (6/fr)
INFO: ngram_search_fwdflat.c(950): 1117 senones evaluated (45/fr)
INFO: ngram_search_fwdflat.c(952): 255 channels searched (10/fr)
INFO: ngram_search_fwdflat.c(954): 255 words searched (10/fr)
INFO: ngram_search_fwdflat.c(957): 240 word transitions (9/fr)
INFO: ngram_search_fwdflat.c(960): fwdflat 0.00 CPU 0.018 xRT
INFO: ngram_search_fwdflat.c(963): fwdflat 0.01 wall 0.039 xRT
2015-11-01 00:42:53.690 OpenEarsSampleApp[2568:1401562] Speech detected…
2015-11-01 00:42:53.691 OpenEarsSampleApp[2568:1401541] Local callback: Pocketsphinx has detected speech.
INFO: cmn_prior.c(99): cmn_prior_update: from < 9.16 0.33 0.01 0.11 -0.29 -0.31 -0.18 -0.15 -0.05 -0.18 -0.07 -0.09 -0.12 >
INFO: cmn_prior.c(116): cmn_prior_update: to < 9.36 0.25 -0.01 0.11 -0.28 -0.28 -0.19 -0.16 -0.05 -0.18 -0.07 -0.09 -0.12 >
2015-11-01 00:42:55.077 OpenEarsSampleApp[2568:1401562] End of speech detected…
INFO: cmn_prior.c(131): cmn_prior_update: from < 9.36 0.25 -0.01 0.11 -0.28 -0.28 -0.19 -0.16 -0.05 -0.18 -0.07 -0.09 -0.12 >
INFO: cmn_prior.c(149): cmn_prior_update: to < 8.66 0.12 -0.02 0.09 -0.24 -0.25 -0.18 -0.15 -0.05 -0.16 -0.07 -0.08 -0.11 >
2015-11-01 00:42:55.078 OpenEarsSampleApp[2568:1401541] Local callback: Pocketsphinx has detected a second of silence, concluding an utterance.
INFO: ngram_search_fwdtree.c(1553): 2662 words recognized (17/fr)
INFO: ngram_search_fwdtree.c(1555): 30359 senones evaluated (198/fr)
INFO: ngram_search_fwdtree.c(1559): 7565 channels searched (49/fr), 596 1st, 5590 last
INFO: ngram_search_fwdtree.c(1562): 4054 words for which last channels evaluated (26/fr)
INFO: ngram_search_fwdtree.c(1564): 99 candidate words for entering last phone (0/fr)
INFO: ngram_search_fwdtree.c(1567): fwdtree 0.55 CPU 0.360 xRT
INFO: ngram_search_fwdtree.c(1570): fwdtree 3.05 wall 1.995 xRT
INFO: ngram_search_fwdflat.c(302): Utterance vocabulary contains 28 words
INFO: ngram_search_fwdflat.c(948): 1567 words recognized (10/fr)
INFO: ngram_search_fwdflat.c(950): 19893 senones evaluated (130/fr)
INFO: ngram_search_fwdflat.c(952): 6693 channels searched (43/fr)
INFO: ngram_search_fwdflat.c(954): 2989 words searched (19/fr)
INFO: ngram_search_fwdflat.c(957): 2169 word transitions (14/fr)
INFO: ngram_search_fwdflat.c(960): fwdflat 0.08 CPU 0.052 xRT
INFO: ngram_search_fwdflat.c(963): fwdflat 0.08 wall 0.052 xRT
INFO: ngram_search.c(1280): lattice start node <s>.0 end node </s>.121
INFO: ngram_search.c(1306): Eliminated 0 nodes before end node
INFO: ngram_search.c(1411): Lattice has 473 nodes, 3278 links
INFO: ps_lattice.c(1380): Bestpath score: -5985
INFO: ps_lattice.c(1384): Normalizer P(O) = alpha(</s>:121:151) = -339341
INFO: ps_lattice.c(1441): Joint P(O,S) = -391224 P(S|O) = -51883
INFO: ngram_search.c(899): bestpath 0.01 CPU 0.005 xRT
INFO: ngram_search.c(902): bestpath 0.01 wall 0.005 xRT
2015-11-01 00:42:55.176 OpenEarsSampleApp[2568:1401562] Pocketsphinx heard “HOLA” with a score of (-51883) and an utterance ID of 8.
2015-11-01 00:42:55.176 OpenEarsSampleApp[2568:1401541] Flite sending interrupt speech request.
2015-11-01 00:42:55.177 OpenEarsSampleApp[2568:1401541] Local callback: The received hypothesis is HOLA with a score of -51883 and an ID of 8
2015-11-01 00:42:55.178 OpenEarsSampleApp[2568:1401541] I’m running flite
2015-11-01 00:42:55.228 OpenEarsSampleApp[2568:1401541] I’m done running flite and it took 0.050284 seconds
2015-11-01 00:42:55.228 OpenEarsSampleApp[2568:1401541] Flite audio player was nil when referenced so attempting to allocate a new audio player.
2015-11-01 00:42:55.228 OpenEarsSampleApp[2568:1401541] Loading speech data for Flite concluded successfully.
2015-11-01 00:42:55.291 OpenEarsSampleApp[2568:1401541] Flite sending suspend recognition notification.
2015-11-01 00:42:55.293 OpenEarsSampleApp[2568:1401541] Local callback: Flite has started speaking
2015-11-01 00:42:55.297 OpenEarsSampleApp[2568:1401541] Local callback: Pocketsphinx has suspended recognition.
2015-11-01 00:42:55.324 OpenEarsSampleApp[2568:1401562] Speech detected…
2015-11-01 00:42:55.325 OpenEarsSampleApp[2568:1401541] Local callback: Pocketsphinx has detected speech.
2015-11-01 00:42:56.597 OpenEarsSampleApp[2568:1401541] AVAudioPlayer did finish playing with success flag of 1
2015-11-01 00:42:56.749 OpenEarsSampleApp[2568:1401541] Flite sending resume recognition notification.
2015-11-01 00:42:57.250 OpenEarsSampleApp[2568:1401541] Local callback: Flite has finished speaking
2015-11-01 00:42:57.258 OpenEarsSampleApp[2568:1401541] setSecondsOfSilence wasn’t set, using default of 0.700000.
2015-11-01 00:42:57.258 OpenEarsSampleApp[2568:1401541] Local callback: Pocketsphinx has resumed recognition.
INFO: cmn_prior.c(131): cmn_prior_update: from < 8.66 0.12 -0.02 0.09 -0.24 -0.25 -0.18 -0.15 -0.05 -0.16 -0.07 -0.08 -0.11 >
INFO: cmn_prior.c(149): cmn_prior_update: to < 8.83 0.12 -0.03 0.06 -0.25 -0.25 -0.19 -0.15 -0.05 -0.16 -0.06 -0.09 -0.11 >
INFO: ngram_search_fwdtree.c(1553): 396 words recognized (14/fr)
INFO: ngram_search_fwdtree.c(1555): 5107 senones evaluated (182/fr)
INFO: ngram_search_fwdtree.c(1559): 1180 channels searched (42/fr), 96 1st, 828 last
INFO: ngram_search_fwdtree.c(1562): 665 words for which last channels evaluated (23/fr)
INFO: ngram_search_fwdtree.c(1564): 38 candidate words for entering last phone (1/fr)
INFO: ngram_search_fwdtree.c(1567): fwdtree 0.19 CPU 0.691 xRT
INFO: ngram_search_fwdtree.c(1570): fwdtree 2.20 wall 7.862 xRT
INFO: ngram_search_fwdflat.c(302): Utterance vocabulary contains 13 words
INFO: ngram_search_fwdflat.c(948): 267 words recognized (10/fr)
INFO: ngram_search_fwdflat.c(950): 1632 senones evaluated (58/fr)
INFO: ngram_search_fwdflat.c(952): 363 channels searched (12/fr)
INFO: ngram_search_fwdflat.c(954): 363 words searched (12/fr)
INFO: ngram_search_fwdflat.c(957): 350 word transitions (12/fr)
INFO: ngram_search_fwdflat.c(960): fwdflat 0.01 CPU 0.053 xRT
INFO: ngram_search_fwdflat.c(963): fwdflat 0.01 wall 0.047 xRT
[/spoiler]November 2, 2015 at 10:43 am #1027147Halle WinklerPolitepixWelcome,
Sorry you’ve experiencing this issue. I’ve heard one other case of it (although it was in the opposite direction – speech was fine but files were not, weirdly) but I don’t yet know what the cause is. What happens when you raise vadThreshold to 4.4?
Another important thing is that the ‘recognitionScore’ value seems almost random.
Unfortunately, with small vocabularies the scores are extremely uninformative and even with big vocabularies they are extremely influenced by many things in the session besides what you would consider nearness of pronunciation. If you search for “score” in the forum search you’ll see many discussions about how narrow the applicability of scoring is. I wish this was better but there are actually very few cases in which scores can be used – definitely not to get some objective idea of the accuracy across multiple sessions.
If I run the WAV file, will I see the good behavior or the bad behavior? It may be a while before I have the opportunity to do so.
November 2, 2015 at 9:12 pm #1027153lytedesignsParticipantNothing. I changed the value of VadThreshold 4.4 , but gives the same performance. Still showing when words are not said (although detects fewer words than 4.3 VAD ) .
With the wav you can see well the failures, but most of the problems gives voice is live .
There are still particular words that are better, but they are rare.Up to this point we do not know very well to do.
If you need the project as I can go via dropbox ( our project and the modified example of openears ).Thanks!
November 2, 2015 at 9:17 pm #1027154Halle WinklerPolitepixIt may be a while before I can check the WAV, but can you create a full session WAV using SaveThatWave as described in the guide to creating replication cases if you’d like me to check it out? Since that WAV has no silent lead-in, I think it must be a manual recording.
November 4, 2015 at 9:22 am #1027188lytedesignsParticipantHello,
we tested a wav generated SaveThatWave saying these words in Spanish:
CARPETA SALUDO COMIDA CARTEL VENEZUELA ELECCIONES GATO DERECHAIn OpenEars + Rejecto we put the words to search:
@”CABEZA”, @”IZQUIERDA”, @”DERECHA”, @”SOBRE”, @”HOLA”, @”CAMBIAR”, @”RÁPIDO”, @”CIUDADANO”, @”URNA”Still finding words unsaid.
Here is the wav file generated with SaveThatWave from the device:
https://dl.dropboxusercontent.com/u/87410097/Rec_device.wavHere is the wav file recorded with the device soft and exporter then wav with ‘afconvert’ (the same as I went):
https://dl.dropboxusercontent.com/u/87410097/Rec2.wavCan you test it?
Thanksconsole log running the saveThatWave wav as testFile:
[spoiler]
2015-11-04 09:21:07.611 OpenEarsTest[3506:336821] Starting OpenEars logging for OpenEars version 2.04 on 32-bit device (or build): iPhone running iOS version: 8.100000
2015-11-04 09:21:07.863 OpenEarsTest[3506:336821] The word URNA was not found in the dictionary /private/var/mobile/Containers/Bundle/Application/07477015-1B5B-413E-8FB2-1B7AF2039661/OpenEarsTest.app/AcousticModelSpanish.bundle/LanguageModelGeneratorLookupList.text/LanguageModelGeneratorLookupList.text.
2015-11-04 09:21:07.865 OpenEarsTest[3506:336821] Now using the fallback method to look up the word URNA
2015-11-04 09:21:07.865 OpenEarsTest[3506:336821] If this is happening more frequently than you would expect, the most likely cause for it is since you are using the Spanish phonetic lookup dictionary is that your words are not in Spanish or aren’t dictionary words.
2015-11-04 09:21:07.867 OpenEarsTest[3506:336821] I’m done running performDictionaryLookup and it took 0.028449 seconds
2015-11-04 09:21:07.870 OpenEarsTest[3506:336821] I’m done running performDictionaryLookup and it took 0.088410 seconds
2015-11-04 09:21:07.880 OpenEarsTest[3506:336821] Starting dynamic language model generationINFO: cmd_ln.c(702): Parsing command line:
sphinx_lm_convert \
-i /var/mobile/Containers/Data/Application/5C951750-9914-4547-B09D-660F7BBE180D/Library/Caches/NameIWantForMyLanguageModelFiles.arpa \
-o /var/mobile/Containers/Data/Application/5C951750-9914-4547-B09D-660F7BBE180D/Library/Caches/NameIWantForMyLanguageModelFiles.DMPCurrent configuration:
[NAME] [DEFLT] [VALUE]
-case
-debug 0
-help no no
-i /var/mobile/Containers/Data/Application/5C951750-9914-4547-B09D-660F7BBE180D/Library/Caches/NameIWantForMyLanguageModelFiles.arpa
-ifmt
-logbase 1.0001 1.000100e+00
-mmap no no
-o /var/mobile/Containers/Data/Application/5C951750-9914-4547-B09D-660F7BBE180D/Library/Caches/NameIWantForMyLanguageModelFiles.DMP
-ofmtINFO: ngram_model_arpa.c(504): ngrams 1=36, 2=68, 3=34
INFO: ngram_model_arpa.c(137): Reading unigrams
INFO: ngram_model_arpa.c(543): 36 = #unigrams created
INFO: ngram_model_arpa.c(197): Reading bigrams
INFO: ngram_model_arpa.c(561): 68 = #bigrams created
INFO: ngram_model_arpa.c(562): 3 = #prob2 entries
INFO: ngram_model_arpa.c(570): 3 = #bo_wt2 entries
INFO: ngram_model_arpa.c(294): Reading trigrams
INFO: ngram_model_arpa.c(583): 34 = #trigrams created
INFO: ngram_model_arpa.c(584): 2 = #prob3 entries
INFO: ngram_model_dmp.c(518): Building DMP model…
INFO: ngram_model_dmp.c(548): 36 = #unigrams created
INFO: ngram_model_dmp.c(649): 68 = #bigrams created
INFO: ngram_model_dmp.c(650): 3 = #prob2 entries
INFO: ngram_model_dmp.c(657): 3 = #bo_wt2 entries
INFO: ngram_model_dmp.c(661): 34 = #trigrams created
INFO: ngram_model_dmp.c(662): 2 = #prob3 entries
2015-11-04 09:21:07.988 OpenEarsTest[3506:336821] Done creating language model with CMUCLMTK in 0.106777 seconds.
INFO: cmd_ln.c(702): Parsing command line:
sphinx_lm_convert \
-i /var/mobile/Containers/Data/Application/5C951750-9914-4547-B09D-660F7BBE180D/Library/Caches/NameIWantForMyLanguageModelFiles.arpa \
-o /var/mobile/Containers/Data/Application/5C951750-9914-4547-B09D-660F7BBE180D/Library/Caches/NameIWantForMyLanguageModelFiles.DMPCurrent configuration:
[NAME] [DEFLT] [VALUE]
-case
-debug 0
-help no no
-i /var/mobile/Containers/Data/Application/5C951750-9914-4547-B09D-660F7BBE180D/Library/Caches/NameIWantForMyLanguageModelFiles.arpa
-ifmt
-logbase 1.0001 1.000100e+00
-mmap no no
-o /var/mobile/Containers/Data/Application/5C951750-9914-4547-B09D-660F7BBE180D/Library/Caches/NameIWantForMyLanguageModelFiles.DMP
-ofmtINFO: ngram_model_arpa.c(504): ngrams 1=36, 2=68, 3=34
INFO: ngram_model_arpa.c(137): Reading unigrams
INFO: ngram_model_arpa.c(543): 36 = #unigrams created
INFO: ngram_model_arpa.c(197): Reading bigrams
INFO: ngram_model_arpa.c(561): 68 = #bigrams created
INFO: ngram_model_arpa.c(562): 3 = #prob2 entries
INFO: ngram_model_arpa.c(570): 3 = #bo_wt2 entries
INFO: ngram_model_arpa.c(294): Reading trigrams
INFO: ngram_model_arpa.c(583): 34 = #trigrams created
INFO: ngram_model_arpa.c(584): 2 = #prob3 entries
INFO: ngram_model_dmp.c(518): Building DMP model…
INFO: ngram_model_dmp.c(548): 36 = #unigrams created
INFO: ngram_model_dmp.c(649): 68 = #bigrams created
INFO: ngram_model_dmp.c(650): 3 = #prob2 entries
INFO: ngram_model_dmp.c(657): 3 = #bo_wt2 entries
INFO: ngram_model_dmp.c(661): 34 = #trigrams created
INFO: ngram_model_dmp.c(662): 2 = #prob3 entries
2015-11-04 09:21:08.000 OpenEarsTest[3506:336821] I’m done running dynamic language model generation and it took 0.356924 seconds
2015-11-04 09:21:08.012 OpenEarsTest[3506:336821] User gave mic permission for this app.
2015-11-04 09:21:08.055 OpenEarsTest[3506:336821] Attempting to start listening session from startRealtimeListeningWithLanguageModelAtPath:
2015-11-04 09:21:08.059 OpenEarsTest[3506:336821] User gave mic permission for this app.
2015-11-04 09:21:08.060 OpenEarsTest[3506:336821] Valid setSecondsOfSilence value of 1.700000 will be used.
2015-11-04 09:21:08.062 OpenEarsTest[3506:336821] Successfully started listening session from startRealtimeListeningWithLanguageModelAtPath:
2015-11-04 09:21:08.064 OpenEarsTest[3506:336888] Starting listening.
2015-11-04 09:21:08.066 OpenEarsTest[3506:336888] about to set up audio session
2015-11-04 09:21:08.068 OpenEarsTest[3506:336888] Creating audio session with default settings.
2015-11-04 09:21:08.178 OpenEarsTest[3506:336907] Audio route has changed for the following reason:
2015-11-04 09:21:08.547 OpenEarsTest[3506:336907] There was a category change. The new category is AVAudioSessionCategoryPlayAndRecord
2015-11-04 09:21:08.564 OpenEarsTest[3506:336907] This is not a case in which OpenEars notifies of a route change. At the close of this function, the new audio route is —SpeakerMicrophoneBuiltIn—. The previous route before changing to this route was <AVAudioSessionRouteDescription: 0x16593b00,
inputs = (null);
outputs = (
“<AVAudioSessionPortDescription: 0x165a7f40, type = Speaker; name = Altavoz; UID = Built-In Speaker; selectedDataSource = (null)>”
)>.
2015-11-04 09:21:08.575 OpenEarsTest[3506:336888] done starting audio unit
INFO: cmd_ln.c(702): Parsing command line:
\
-lm /var/mobile/Containers/Data/Application/5C951750-9914-4547-B09D-660F7BBE180D/Library/Caches/NameIWantForMyLanguageModelFiles.DMP \
-vad_prespeech 10 \
-vad_postspeech 170 \
-vad_threshold 4.300000 \
-remove_noise yes \
-remove_silence yes \
-bestpath no \
-lw 6.500000 \
-dict /var/mobile/Containers/Data/Application/5C951750-9914-4547-B09D-660F7BBE180D/Library/Caches/NameIWantForMyLanguageModelFiles.dic \
-hmm /private/var/mobile/Containers/Bundle/Application/07477015-1B5B-413E-8FB2-1B7AF2039661/OpenEarsTest.app/AcousticModelSpanish.bundleCurrent configuration:
[NAME] [DEFLT] [VALUE]
-agc none none
-agcthresh 2.0 2.000000e+00
-allphone
-allphone_ci no no
-alpha 0.97 9.700000e-01
-argfile
-ascale 20.0 2.000000e+01
-aw 1 1
-backtrace no no
-beam 1e-48 1.000000e-48
-bestpath yes no
-bestpathlw 9.5 9.500000e+00
-ceplen 13 13
-cmn current current
-cmninit 8.0 8.0
-compallsen no no
-debug 0
-dict /var/mobile/Containers/Data/Application/5C951750-9914-4547-B09D-660F7BBE180D/Library/Caches/NameIWantForMyLanguageModelFiles.dic
-dictcase no no
-dither no no
-doublebw no no
-ds 1 1
-fdict
-feat 1s_c_d_dd 1s_c_d_dd
-featparams
-fillprob 1e-8 1.000000e-08
-frate 100 100
-fsg
-fsgusealtpron yes yes
-fsgusefiller yes yes
-fwdflat yes yes
-fwdflatbeam 1e-64 1.000000e-64
-fwdflatefwid 4 4
-fwdflatlw 8.5 8.500000e+00
-fwdflatsfwin 25 25
-fwdflatwbeam 7e-29 7.000000e-29
-fwdtree yes yes
-hmm /private/var/mobile/Containers/Bundle/Application/07477015-1B5B-413E-8FB2-1B7AF2039661/OpenEarsTest.app/AcousticModelSpanish.bundle
-input_endian little little
-jsgf
-keyphrase
-kws
-kws_plp 1e-1 1.000000e-01
-kws_threshold 1 1.000000e+00
-latsize 5000 5000
-lda
-ldadim 0 0
-lifter 0 0
-lm /var/mobile/Containers/Data/Application/5C951750-9914-4547-B09D-660F7BBE180D/Library/Caches/NameIWantForMyLanguageModelFiles.DMP
-lmctl
-lmname
-logbase 1.0001 1.000100e+00
-logfn
-logspec no no
-lowerf 133.33334 1.333333e+02
-lpbeam 1e-40 1.000000e-40
-lponlybeam 7e-29 7.000000e-29
-lw 6.5 6.500000e+00
-maxhmmpf 30000 30000
-maxwpf -1 -1
-mdef
-mean
-mfclogdir
-min_endfr 0 0
-mixw
-mixwfloor 0.0000001 1.000000e-07
-mllr
-mmap yes yes
-ncep 13 13
-nfft 512 512
-nfilt 40 40
-nwpen 1.0 1.000000e+00
-pbeam 1e-48 1.000000e-48
-pip 1.0 1.000000e+00
-pl_beam 1e-10 1.000000e-10
-pl_pbeam 1e-10 1.000000e-10
-pl_pip 1.0 1.000000e+00
-pl_weight 3.0 3.000000e+00
-pl_window 5 5
-rawlogdir
-remove_dc no no
-remove_noise yes yes
-remove_silence yes yes
-round_filters yes yes
-samprate 16000 1.600000e+04
-seed -1 -1
-sendump
-senlogdir
-senmgau
-silprob 0.005 5.000000e-03
-smoothspec no no
-svspec
-tmat
-tmatfloor 0.0001 1.000000e-04
-topn 4 4
-topn_beam 0 0
-toprule
-transform legacy legacy
-unit_area yes yes
-upperf 6855.4976 6.855498e+03
-uw 1.0 1.000000e+00
-vad_postspeech 50 170
-vad_prespeech 10 10
-vad_threshold 2.0 4.300000e+00
-var
-varfloor 0.0001 1.000000e-04
-varnorm no no
-verbose no no
-warp_params
-warp_type inverse_linear inverse_linear
-wbeam 7e-29 7.000000e-29
-wip 0.65 6.500000e-01
-wlen 0.025625 2.562500e-02INFO: cmd_ln.c(702): Parsing command line:
\
-feat s3_1x39Current configuration:
[NAME] [DEFLT] [VALUE]
-agc none none
-agcthresh 2.0 2.000000e+00
-alpha 0.97 9.700000e-01
-ceplen 13 13
-cmn current current
-cmninit 8.0 8.0
-dither no no
-doublebw no no
-feat 1s_c_d_dd s3_1x39
-frate 100 100
-input_endian little little
-lda
-ldadim 0 0
-lifter 0 0
-logspec no no
-lowerf 133.33334 1.333333e+02
-ncep 13 13
-nfft 512 512
-nfilt 40 40
-remove_dc no no
-remove_noise yes yes
-remove_silence yes yes
-round_filters yes yes
-samprate 16000 1.600000e+04
-seed -1 -1
-smoothspec no no
-svspec
-transform legacy legacy
-unit_area yes yes
-upperf 6855.4976 6.855498e+03
-vad_postspeech 50 170
-vad_prespeech 10 10
-vad_threshold 2.0 4.300000e+00
-varnorm no no
-verbose no no
-warp_params
-warp_type inverse_linear inverse_linear
-wlen 0.025625 2.562500e-02INFO: acmod.c(252): Parsed model-specific feature parameters from /private/var/mobile/Containers/Bundle/Application/07477015-1B5B-413E-8FB2-1B7AF2039661/OpenEarsTest.app/AcousticModelSpanish.bundle/feat.params
INFO: feat.c(715): Initializing feature stream to type: ‘s3_1x39′, ceplen=13, CMN=’current’, VARNORM=’no’, AGC=’none’
INFO: cmn.c(143): mean[0]= 12.00, mean[1..12]= 0.0
INFO: mdef.c(518): Reading model definition: /private/var/mobile/Containers/Bundle/Application/07477015-1B5B-413E-8FB2-1B7AF2039661/OpenEarsTest.app/AcousticModelSpanish.bundle/mdef
INFO: bin_mdef.c(181): Allocating 27954 * 8 bytes (218 KiB) for CD tree
INFO: tmat.c(206): Reading HMM transition probability matrices: /private/var/mobile/Containers/Bundle/Application/07477015-1B5B-413E-8FB2-1B7AF2039661/OpenEarsTest.app/AcousticModelSpanish.bundle/transition_matrices
INFO: acmod.c(124): Attempting to use PTM computation module
INFO: ms_gauden.c(198): Reading mixture gaussian parameter: /private/var/mobile/Containers/Bundle/Application/07477015-1B5B-413E-8FB2-1B7AF2039661/OpenEarsTest.app/AcousticModelSpanish.bundle/means
INFO: ms_gauden.c(292): 2630 codebook, 1 feature, size:
INFO: ms_gauden.c(294): 16×39
INFO: ms_gauden.c(198): Reading mixture gaussian parameter: /private/var/mobile/Containers/Bundle/Application/07477015-1B5B-413E-8FB2-1B7AF2039661/OpenEarsTest.app/AcousticModelSpanish.bundle/variances
INFO: ms_gauden.c(292): 2630 codebook, 1 feature, size:
INFO: ms_gauden.c(294): 16×39
INFO: ms_gauden.c(354): 16 variance values floored
INFO: ptm_mgau.c(801): Number of codebooks exceeds 256: 2630
INFO: acmod.c(126): Attempting to use semi-continuous computation module
INFO: ms_gauden.c(198): Reading mixture gaussian parameter: /private/var/mobile/Containers/Bundle/Application/07477015-1B5B-413E-8FB2-1B7AF2039661/OpenEarsTest.app/AcousticModelSpanish.bundle/means
INFO: ms_gauden.c(292): 2630 codebook, 1 feature, size:
INFO: ms_gauden.c(294): 16×39
INFO: ms_gauden.c(198): Reading mixture gaussian parameter: /private/var/mobile/Containers/Bundle/Application/07477015-1B5B-413E-8FB2-1B7AF2039661/OpenEarsTest.app/AcousticModelSpanish.bundle/variances
INFO: ms_gauden.c(292): 2630 codebook, 1 feature, size:
INFO: ms_gauden.c(294): 16×39
INFO: ms_gauden.c(354): 16 variance values floored
INFO: acmod.c(128): Falling back to general multi-stream GMM computation
INFO: ms_gauden.c(198): Reading mixture gaussian parameter: /private/var/mobile/Containers/Bundle/Application/07477015-1B5B-413E-8FB2-1B7AF2039661/OpenEarsTest.app/AcousticModelSpanish.bundle/means
INFO: ms_gauden.c(292): 2630 codebook, 1 feature, size:
INFO: ms_gauden.c(294): 16×39
INFO: ms_gauden.c(198): Reading mixture gaussian parameter: /private/var/mobile/Containers/Bundle/Application/07477015-1B5B-413E-8FB2-1B7AF2039661/OpenEarsTest.app/AcousticModelSpanish.bundle/variances
INFO: ms_gauden.c(292): 2630 codebook, 1 feature, size:
INFO: ms_gauden.c(294): 16×39
INFO: ms_gauden.c(354): 16 variance values floored
INFO: ms_senone.c(149): Reading senone mixture weights: /private/var/mobile/Containers/Bundle/Application/07477015-1B5B-413E-8FB2-1B7AF2039661/OpenEarsTest.app/AcousticModelSpanish.bundle/mixture_weights
INFO: ms_senone.c(200): Truncating senone logs3(pdf) values by 10 bits
INFO: ms_senone.c(207): Not transposing mixture weights in memory
INFO: ms_senone.c(268): Read mixture weights for 2630 senones: 1 features x 16 codewords
INFO: ms_senone.c(320): Mapping senones to individual codebooks
INFO: ms_mgau.c(141): The value of topn: 4
INFO: phone_loop_search.c(115): State beam -225 Phone exit beam -225 Insertion penalty 0
INFO: dict.c(320): Allocating 4130 * 20 bytes (80 KiB) for word entries
INFO: dict.c(333): Reading main dictionary: /var/mobile/Containers/Data/Application/5C951750-9914-4547-B09D-660F7BBE180D/Library/Caches/NameIWantForMyLanguageModelFiles.dic
INFO: dict.c(213): Allocated 0 KiB for strings, 0 KiB for phones
INFO: dict.c(336): 34 words read
INFO: dict2pid.c(396): Building PID tables for dictionary
INFO: dict2pid.c(406): Allocating 26^3 * 2 bytes (34 KiB) for word-initial triphones
INFO: dict2pid.c(132): Allocated 8216 bytes (8 KiB) for word-final triphones
INFO: dict2pid.c(196): Allocated 8216 bytes (8 KiB) for single-phone word triphones
INFO: ngram_model_arpa.c(79): No \data\ mark in LM file
INFO: ngram_model_dmp.c(166): Will use memory-mapped I/O for LM file
INFO: ngram_model_dmp.c(220): ngrams 1=36, 2=68, 3=34
INFO: ngram_model_dmp.c(266): 36 = LM.unigrams(+trailer) read
INFO: ngram_model_dmp.c(312): 68 = LM.bigrams(+trailer) read
INFO: ngram_model_dmp.c(338): 34 = LM.trigrams read
INFO: ngram_model_dmp.c(363): 3 = LM.prob2 entries read
INFO: ngram_model_dmp.c(383): 3 = LM.bo_wt2 entries read
INFO: ngram_model_dmp.c(403): 2 = LM.prob3 entries read
INFO: ngram_model_dmp.c(431): 1 = LM.tseg_base entries read
INFO: ngram_model_dmp.c(487): 36 = ascii word strings read
INFO: ngram_search_fwdtree.c(99): 8 unique initial diphones
INFO: ngram_search_fwdtree.c(148): 0 root, 0 non-root channels, 29 single-phone words
INFO: ngram_search_fwdtree.c(186): Creating search tree
INFO: ngram_search_fwdtree.c(192): before: 0 root, 0 non-root channels, 29 single-phone words
INFO: ngram_search_fwdtree.c(326): after: max nonroot chan increased to 164
INFO: ngram_search_fwdtree.c(339): after: 8 root, 36 non-root channels, 28 single-phone words
INFO: ngram_search_fwdflat.c(157): fwdflat: min_ef_width = 4, max_sf_win = 25
2015-11-04 09:21:12.627 OpenEarsTest[3506:336888] Listening.
2015-11-04 09:21:12.631 OpenEarsTest[3506:336888] Project has these words or phrases in its dictionary:
___REJ_Y
___REJ_X
___REJ_V
___REJ_U
___REJ_T
___REJ_S
___REJ_RR
___REJ_R
___REJ_P
___REJ_O
___REJ_N
___REJ_M
___REJ_LL
___REJ_L
___REJ_K
___REJ_J
___REJ_I
___REJ_GN
___REJ_G
___REJ_F
___REJ_E
___REJ_D
___REJ_CH
___REJ_B
___REJ_A
CABEZA
CAMBIAR
CIUDADANO
DERECHA
HOLA
IZQUIERDA
…and 4 more.
2015-11-04 09:21:12.632 OpenEarsTest[3506:336888] Recognition loop has started
2015-11-04 09:21:12.637 OpenEarsTest[3506:336821] Pocketsphinx is now listening.
2015-11-04 09:21:14.284 OpenEarsTest[3506:336888] Speech detected…
2015-11-04 09:21:14.285 OpenEarsTest[3506:336821] Pocketsphinx has detected speech.
2015-11-04 09:21:14.287 OpenEarsTest[3506:336886] Pocketsphinx heard ” ” with a score of (-843) and an utterance ID of 0.
2015-11-04 09:21:14.288 OpenEarsTest[3506:336886] Hypothesis was null so we aren’t returning it. If you want null hypotheses to also be returned, set OEPocketsphinxController’s property returnNullHypotheses to TRUE before starting OEPocketsphinxController.
2015-11-04 09:21:14.471 OpenEarsTest[3506:336888] Pocketsphinx heard “HOLA” with a score of (-1487) and an utterance ID of 1.
2015-11-04 09:21:14.472 OpenEarsTest[3506:336821] rapidEarsDidReceiveLiveSpeechHypothesis: The received hypothesis is HOLA with a score of -1487
2015-11-04 09:21:14.716 OpenEarsTest[3506:336886] Pocketsphinx heard ” ” with a score of (-2476) and an utterance ID of 2.
2015-11-04 09:21:14.717 OpenEarsTest[3506:336886] Hypothesis was null so we aren’t returning it. If you want null hypotheses to also be returned, set OEPocketsphinxController’s property returnNullHypotheses to TRUE before starting OEPocketsphinxController.
2015-11-04 09:21:14.926 OpenEarsTest[3506:336886] Pocketsphinx heard ” ” with a score of (-4259) and an utterance ID of 3.
2015-11-04 09:21:15.130 OpenEarsTest[3506:336886] Pocketsphinx heard ” ” with a score of (-5109) and an utterance ID of 4.
2015-11-04 09:21:15.326 OpenEarsTest[3506:336886] Pocketsphinx heard ” ” with a score of (-5523) and an utterance ID of 5.
2015-11-04 09:21:15.632 OpenEarsTest[3506:336886] Pocketsphinx heard ” ” with a score of (-5767) and an utterance ID of 6.
2015-11-04 09:21:15.901 OpenEarsTest[3506:336886] Pocketsphinx heard ” ” with a score of (-6465) and an utterance ID of 7.
2015-11-04 09:21:16.166 OpenEarsTest[3506:336886] Pocketsphinx heard “HOLA” with a score of (-8759) and an utterance ID of 8.
2015-11-04 09:21:16.167 OpenEarsTest[3506:336821] rapidEarsDidReceiveLiveSpeechHypothesis: The received hypothesis is HOLA with a score of -8759
2015-11-04 09:21:16.458 OpenEarsTest[3506:336886] Pocketsphinx heard “URNA HOLA” with a score of (-10135) and an utterance ID of 9.
2015-11-04 09:21:16.459 OpenEarsTest[3506:336821] rapidEarsDidReceiveLiveSpeechHypothesis: The received hypothesis is URNA HOLA with a score of -10135
2015-11-04 09:21:16.754 OpenEarsTest[3506:336886] Pocketsphinx heard “URNA HOLA” with a score of (-10189) and an utterance ID of 10.
2015-11-04 09:21:16.986 OpenEarsTest[3506:336886] Pocketsphinx heard “URNA HOLA” with a score of (-11234) and an utterance ID of 11.
2015-11-04 09:21:17.191 OpenEarsTest[3506:336886] Pocketsphinx heard “URNA HOLA” with a score of (-12147) and an utterance ID of 12.
2015-11-04 09:21:17.477 OpenEarsTest[3506:336886] Pocketsphinx heard “URNA HOLA” with a score of (-14526) and an utterance ID of 13.
2015-11-04 09:21:17.752 OpenEarsTest[3506:336886] Pocketsphinx heard “URNA HOLA DERECHA” with a score of (-15029) and an utterance ID of 14.
2015-11-04 09:21:17.753 OpenEarsTest[3506:336821] rapidEarsDidReceiveLiveSpeechHypothesis: The received hypothesis is URNA HOLA DERECHA with a score of -15029
2015-11-04 09:21:18.047 OpenEarsTest[3506:336886] Pocketsphinx heard “URNA HOLA DERECHA” with a score of (-16053) and an utterance ID of 15.
2015-11-04 09:21:18.319 OpenEarsTest[3506:336886] Pocketsphinx heard “URNA HOLA DERECHA” with a score of (-16390) and an utterance ID of 16.
INFO: ngram_search.c(462): Resized backpointer table to 10000 entries
2015-11-04 09:21:18.614 OpenEarsTest[3506:336886] Pocketsphinx heard “URNA HOLA DERECHA” with a score of (-16724) and an utterance ID of 17.
2015-11-04 09:21:18.898 OpenEarsTest[3506:336886] Pocketsphinx heard “URNA HOLA DERECHA” with a score of (-17003) and an utterance ID of 18.
2015-11-04 09:21:19.192 OpenEarsTest[3506:336886] Pocketsphinx heard “URNA HOLA DERECHA” with a score of (-17244) and an utterance ID of 19.
2015-11-04 09:21:19.256 OpenEarsTest[3506:336888] End of speech detected…
2015-11-04 09:21:19.257 OpenEarsTest[3506:336821] Pocketsphinx has detected a period of silence, concluding an utterance.
INFO: cmn_prior.c(131): cmn_prior_update: from < 8.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 >
INFO: cmn_prior.c(149): cmn_prior_update: to < 8.89 0.07 -0.36 0.01 -0.16 -0.01 -0.26 -0.21 -0.05 -0.18 -0.09 -0.07 -0.07 >
INFO: ngram_search_fwdtree.c(1553): 6748 words recognized (14/fr)
INFO: ngram_search_fwdtree.c(1555): 119026 senones evaluated (245/fr)
INFO: ngram_search_fwdtree.c(1559): 28716 channels searched (59/fr), 3815 1st, 18042 last
INFO: ngram_search_fwdtree.c(1562): 12032 words for which last channels evaluated (24/fr)
INFO: ngram_search_fwdtree.c(1564): 297 candidate words for entering last phone (0/fr)
INFO: ngram_search_fwdtree.c(1567): fwdtree 4.81 CPU 0.990 xRT
INFO: ngram_search_fwdtree.c(1570): fwdtree 6.56 wall 1.350 xRT
INFO: ngram_search_fwdflat.c(302): Utterance vocabulary contains 31 words
INFO: ngram_search_fwdflat.c(945): 3539 words recognized (7/fr)
INFO: ngram_search_fwdflat.c(947): 77412 senones evaluated (159/fr)
INFO: ngram_search_fwdflat.c(949): 22670 channels searched (46/fr)
INFO: ngram_search_fwdflat.c(951): 7729 words searched (15/fr)
INFO: ngram_search_fwdflat.c(954): 6070 word transitions (12/fr)
INFO: ngram_search_fwdflat.c(957): fwdflat 1.97 CPU 0.406 xRT
INFO: ngram_search_fwdflat.c(960): fwdflat 2.07 wall 0.425 xRT
2015-11-04 09:21:21.358 OpenEarsTest[3506:336888] Pocketsphinx heard “URNA HOLA DERECHA” with a score of (-17696) and an utterance ID of 20.
2015-11-04 09:21:21.360 OpenEarsTest[3506:336821] rapidEarsDidReceiveFinishedSpeechHypothesis: The received hypothesis is URNA HOLA DERECHA with a score of -17696
[/spoiler]November 4, 2015 at 9:46 am #1027190Halle WinklerPolitepixHi,
Here is the wav file generated with SaveThatWave from the device
This doesn’t seem to be a file generated from SaveThatWave’s startSessionDebugRecord method. Can you check out the documentation about that method and make sure that you are sending the unaltered results of its output as found in the caches directory of the device after the entire app session has completed?
November 4, 2015 at 1:44 pm #1027191lytedesignsParticipantOk, I started with startSessionDebugRecord and this is generated wav:
https://dl.dropboxusercontent.com/u/87410097/Rec_device.wavThe problem remains the same. As much as you save in one way or another, the recognition problem is still there.
Thank you and hope to use your soft as we like a lot, if not we will have to try other alternatives.November 4, 2015 at 4:47 pm #1027192Halle WinklerPolitepixAs much as you save in one way or another, the recognition problem is still there.
Perhaps (my understanding is that you started this issue report by saying the issue was not seriously occurring in recordings), but if you give me a SaveThatWave file, I can review the actual OpenEars session and what OpenEars recorded using its audio driver. On the other hand, if you send me a file that was recorded with a different app, I won’t know anything about what OpenEars’ audio driver did in the session. And if you don’t accurately tell me how an audio file originates, I will spend a lot of time looking into something without knowing what it is, which wouldn’t bring much. It’s going to take a while to get to review this while the next update is released, so if this is a showstopper for you and you have time pressure, check into the alternatives as well.
November 5, 2015 at 3:27 pm #1027202TheLionParticipantHello Halle,
First I want to thank you for your time.
It can be a compatibility problem with iOS 9 and iOS 9.1?
We followed these steeps:
1- We have downloaded your original example from http://cdn1.politepix.com/wp-content/uploads/OpenEarsDistribution.tar.bz2
2- Without modifying anything on the project, only deactivating the bitcode and then compiling the Application, it doesn`t work correctly. It detects all the word and sound (ambient sound, car, dog bark…) as valid words like it happens right, left… randomly.
We have test it in: iPhone 6 plus, iPhone 6s, iPhone 6 and iPhone 5s and the effect is the same in all platforms.
Thank you very much.
HugoNovember 5, 2015 at 3:36 pm #1027206Halle WinklerPolitepixHello,
Please take a look at the FAQ and maybe read a bit in the forums about the out of vocabulary problem and noise sensitivity. Here is a specific topic to get started with, but there is more in the FAQ and elsewhere:
If you want to discuss it further, please be so kind as to start your own topic for it so it isn’t intermixed with a different issue that is in the process of discussion in this topic, since that conversation would be difficult to conduct effectively, thank you.
November 5, 2015 at 4:20 pm #1027213lytedesignsParticipantHello, im try to reply but the system show a error “Your reply cannot be created at this time”.
I attach the txt with all logs and original reply in dropbox.
https://dl.dropboxusercontent.com/u/87410097/last_post_with_reply_problems.rtfSorry, my English is very bad and maybe we are not well understood.
The main problem is that if we have a word to search and tell others words, openears recognizes words unsaid. Say “CABEZA” and recognized “HOLA” for example.
The last wav file that you spent was recorded using SaveThatWave plugin executing startSessionDebugRecord with the app running and talking on the device.
https://dl.dropboxusercontent.com/u/87410097/Rec_device.wav
(this wav is generated with SaveThatWave on device)I’m sorry because I forgot to attach the log in the previous message. I imagine it will be necessary.
Now i attach the logs when the session was recorded wav, and when I tried to play as Testfile.Also if need, i can attach the ultra simple project with the test. Ready to run.
November 10, 2015 at 7:03 pm #1027247lytedesignsParticipantHello, could you prove or see something with wav or with the logs?
I think I’ve understood that an update will come up soon. It is correct?
Do you think that this new version would work best (in our case)?Thanks again!
November 10, 2015 at 9:05 pm #1027249Halle WinklerPolitepixHello,
Sorry, the next update isn’t related to anything which might be involved in this issue. It may be a while before I can check your examples.
-
AuthorPosts
- You must be logged in to reply to this topic.