[Resolved] Flite and RapidEars

[adb]

What is the correct way to use fliteController with RapidEars for feedback on hypothesis? If I try to have the fliteController give feedback in rapidEarsDidReceiveFinishedSpeechHypothesis: but it gets picked up by RapidEars. I need a reliable way to suspend and resume while flite is speaking. I have tried to suspend/resume pocketsphynx to no avail.

[Halle Winkler]

Hello,

FliteController handles its own suspend and resume, you don’t need to do it.

[adb]

I read that as well but it is not working. RapidEars picks up flite in rapidEarsDidReceiveFinishedSpeechHypothesis: after it is said.

[Halle Winkler]

OK, can you check out this topic about logging and show the complete logging (OpenEarsLogging, verbosePocketsphinx, setRapidEarsToVerbose:TRUE) from a session in which that happens:

Please read before you post – how to troubleshoot and provide logging info here

[adb]

Halle,

Thanks for the quick responses. Here is the log data:

2014-11-20 15:23:26.169 TimeFlowV2[752:104140] Starting OpenEars logging for OpenEars version 1.7 on 32-bit device: iPhone running iOS version: 8.100000
2014-11-20 15:23:26.321 TimeFlowV2[752:104140] I’m done running performDictionaryLookup and it took 0.075714 seconds
2014-11-20 15:23:26.426 TimeFlowV2[752:104140] Starting dynamic language model generation
2014-11-20 15:23:26.432 TimeFlowV2[752:104140] Able to open /var/mobile/Containers/Data/Application/7314BE0C-59C4-46C3-ACD2-13851947528E/Library/Caches/TimeFlowGrammerRuleSet.corpus for reading
2014-11-20 15:23:26.433 TimeFlowV2[752:104140] Able to open /var/mobile/Containers/Data/Application/7314BE0C-59C4-46C3-ACD2-13851947528E/Library/Caches/TimeFlowGrammerRuleSet_pipe.txt for writing
2014-11-20 15:23:26.434 TimeFlowV2[752:104140] Starting text2wfreq_impl
2014-11-20 15:23:26.449 TimeFlowV2[752:104140] Done with text2wfreq_impl
2014-11-20 15:23:26.450 TimeFlowV2[752:104140] Able to open /var/mobile/Containers/Data/Application/7314BE0C-59C4-46C3-ACD2-13851947528E/Library/Caches/TimeFlowGrammerRuleSet_pipe.txt for reading.
2014-11-20 15:23:26.452 TimeFlowV2[752:104140] Able to open /var/mobile/Containers/Data/Application/7314BE0C-59C4-46C3-ACD2-13851947528E/Library/Caches/TimeFlowGrammerRuleSet.vocab for reading.
2014-11-20 15:23:26.452 TimeFlowV2[752:104140] Starting wfreq2vocab
2014-11-20 15:23:26.454 TimeFlowV2[752:104140] Done with wfreq2vocab
2014-11-20 15:23:26.455 TimeFlowV2[752:104140] Starting text2idngram
2014-11-20 15:23:26.475 TimeFlowV2[752:104140] Done with text2idngram
2014-11-20 15:23:26.480 TimeFlowV2[752:104140] Starting idngram2lm

2014-11-20 15:23:26.491 TimeFlowV2[752:104140] Done with idngram2lm
2014-11-20 15:23:26.492 TimeFlowV2[752:104140] Starting sphinx_lm_convert
2014-11-20 15:23:26.497 TimeFlowV2[752:104140] Finishing sphinx_lm_convert
2014-11-20 15:23:26.502 TimeFlowV2[752:104140] Done creating language model with CMUCLMTK in 0.074842 seconds.
2014-11-20 15:23:26.508 TimeFlowV2[752:104140] Generating fast grammar took 0.300853 seconds
2014-11-20 15:23:26.510 TimeFlowV2[752:104140] Starting sphinx_lm_convert
2014-11-20 15:23:26.516 TimeFlowV2[752:104140] Finishing sphinx_lm_convert

2014-11-20 15:23:31.763 TimeFlowV2[752:104140] Starting dynamic language model generation
2014-11-20 15:23:31.772 TimeFlowV2[752:104140] Able to open /var/mobile/Containers/Data/Application/7314BE0C-59C4-46C3-ACD2-13851947528E/Library/Caches/TimeFlowGrammerRuleSet2.corpus for reading
2014-11-20 15:23:31.773 TimeFlowV2[752:104140] Able to open /var/mobile/Containers/Data/Application/7314BE0C-59C4-46C3-ACD2-13851947528E/Library/Caches/TimeFlowGrammerRuleSet2_pipe.txt for writing
2014-11-20 15:23:31.774 TimeFlowV2[752:104140] Starting text2wfreq_impl
2014-11-20 15:23:31.795 TimeFlowV2[752:104140] Done with text2wfreq_impl
2014-11-20 15:23:31.796 TimeFlowV2[752:104140] Able to open /var/mobile/Containers/Data/Application/7314BE0C-59C4-46C3-ACD2-13851947528E/Library/Caches/TimeFlowGrammerRuleSet2_pipe.txt for reading.
2014-11-20 15:23:31.798 TimeFlowV2[752:104140] Able to open /var/mobile/Containers/Data/Application/7314BE0C-59C4-46C3-ACD2-13851947528E/Library/Caches/TimeFlowGrammerRuleSet2.vocab for reading.
2014-11-20 15:23:31.798 TimeFlowV2[752:104140] Starting wfreq2vocab
2014-11-20 15:23:31.805 TimeFlowV2[752:104140] Done with wfreq2vocab
2014-11-20 15:23:31.806 TimeFlowV2[752:104140] Starting text2idngram
2014-11-20 15:23:31.841 TimeFlowV2[752:104140] Done with text2idngram
2014-11-20 15:23:31.845 TimeFlowV2[752:104140] Starting idngram2lm

2014-11-20 15:23:31.874 TimeFlowV2[752:104140] Done with idngram2lm
2014-11-20 15:23:31.875 TimeFlowV2[752:104140] Starting sphinx_lm_convert
2014-11-20 15:23:31.890 TimeFlowV2[752:104140] Finishing sphinx_lm_convert
2014-11-20 15:23:31.895 TimeFlowV2[752:104140] Done creating language model with CMUCLMTK in 0.131294 seconds.
2014-11-20 15:23:31.911 TimeFlowV2[752:104140] Generating fast grammar took 5.379857 seconds
2014-11-20 15:23:31.912 TimeFlowV2[752:104140] Starting sphinx_lm_convert
2014-11-20 15:23:31.926 TimeFlowV2[752:104140] Finishing sphinx_lm_convert
2014-11-20 15:23:31.930 TimeFlowV2[752:104140] Accuracy set to 20.
2014-11-20 15:23:31.931 TimeFlowV2[752:104140] User gave mic permission for this app.
2014-11-20 15:23:31.932 TimeFlowV2[752:104140] A sample rate was requested that isn’t one of the two supported values of 16000 or 8000 so we will use the default of 16000.
2014-11-20 15:23:31.933 TimeFlowV2[752:104140] The audio session has already been initialized but we will override its properties.
2014-11-20 15:23:31.933 TimeFlowV2[752:104140] Checking and resetting all audio session settings.
2014-11-20 15:23:31.935 TimeFlowV2[752:104140] audioCategory is incorrect, we will change it.
2014-11-20 15:23:31.935 TimeFlowV2[752:104140] audioCategory is now on the correct setting of kAudioSessionCategory_PlayAndRecord.
2014-11-20 15:23:31.936 TimeFlowV2[752:104140] bluetoothInput is incorrect, we will change it.
2014-11-20 15:23:31.936 TimeFlowV2[752:104140] bluetooth input is now on the correct setting of 1.
2014-11-20 15:23:31.939 TimeFlowV2[752:104140] Output Device: ReceiverAndMicrophone.
2014-11-20 15:23:31.939 TimeFlowV2[752:104140] categoryDefaultToSpeaker is incorrect, we will change it.
2014-11-20 15:23:31.940 TimeFlowV2[752:104140] CategoryDefaultToSpeaker is now on the correct setting of 1.
2014-11-20 15:23:31.941 TimeFlowV2[752:104140] preferredBufferSize is incorrect, we will change it.
2014-11-20 15:23:31.942 TimeFlowV2[752:104140] PreferredBufferSize is now on the correct setting of 0.128000.
2014-11-20 15:23:31.942 TimeFlowV2[752:104140] preferredSampleRateCheck is incorrect, we will change it.
2014-11-20 15:23:31.943 TimeFlowV2[752:104140] preferred hardware sample rate is now on the correct setting of 16000.000000.
2014-11-20 15:23:32.232 TimeFlowV2[752:104140] AudioSessionManager startAudioSession has reached the end of the initialization.
2014-11-20 15:23:32.233 TimeFlowV2[752:104140] Exiting startAudioSession.
2014-11-20 15:23:32.248 TimeFlowV2[752:104246] setSecondsOfSilence value of 0.000000 was too large or too small or was NULL, using default of 0.700000.
2014-11-20 15:23:32.252 TimeFlowV2[752:104246] Project has these words or phrases in its dictionary:
OMMITTED
2014-11-20 15:23:32.253 TimeFlowV2[752:104246] Recognition loop has started
INFO: file_omitted(0): Parsing command line:
\
-lm /var/mobile/Containers/Data/Application/7314BE0C-59C4-46C3-ACD2-13851947528E/Library/Caches/TimeFlowGrammerRuleSet.DMP \
-beam 1e-66 \
-bestpath yes \
-dict /var/mobile/Containers/Data/Application/7314BE0C-59C4-46C3-ACD2-13851947528E/Library/Caches/TimeFlowGrammerRuleSet.dic \
-hmm /private/var/mobile/Containers/Bundle/Application/764368A1-D88E-4705-A908-103D2213B5D5/TimeFlowV2.app/AcousticModelEnglish.bundle \
-lw 6.5 \
-samprate 16000

Current configuration:
[NAME] [DEFLT] [VALUE]
-adchdr 0 0
-adcin no no
-agc none none
-agcthresh 2.0 2.000000e+00
-align
-alpha 0.97 9.700000e-01
-arcdumpdir
-ascale 20.0 2.000000e+01
-aw 1 1
-backtrace no no
-beam 1e-48 1.000000e-66
-bestpath yes yes
-bestpathlw 9.5 9.500000e+00
-bghist no no
-cepdir
-cepext .mfc .mfc
-ceplen 13 13
-cmn current current
-cmninit 8.0 8.0
-compallsen no no
-ctl
-ctlcount -1 -1
-ctlincr 1 1
-ctloffset 0 0
-debug 0
-dict /var/mobile/Containers/Data/Application/7314BE0C-59C4-46C3-ACD2-13851947528E/Library/Caches/TimeFlowGrammerRuleSet.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
-fwdflatlw 8.5 8.500000e+00
-fwdflatsfwin 25 25
-fwdflatwbeam 7e-29 7.000000e-29
-fwdtree yes yes
-fwdtreelm
-hmm /private/var/mobile/Containers/Bundle/Application/764368A1-D88E-4705-A908-103D2213B5D5/TimeFlowV2.app/AcousticModelEnglish.bundle
-hyp
-hypprefix
-input_endian little little
-jsgf
-kdmaxbbi -1 -1
-kdmaxdepth 0 0
-kdtree
-latsize 512 512
-lda
-ldadim 0 0
-lextreedump 0 0
-lifter 0 0
-lm /var/mobile/Containers/Data/Application/7314BE0C-59C4-46C3-ACD2-13851947528E/Library/Caches/TimeFlowGrammerRuleSet.DMP
-lmctl
-lmname default default
-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 -1 -1
-maxnewoov 20 20
-maxsilfr -1 -1
-maxwpf 50 50
-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-5 1.000000e-05
-pl_window 0 0
-rawlogdir
-remove_dc no no
-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
-usewdphones no no
-uw 1.0 1.000000e+00
-var
-varfloor 0.0001 1.000000e-04
-varnorm no no
-verbose no no
-vm
-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-02

INFO: file_omitted(0): Parsing command line:
\
-nfilt 20 \
-lowerf 1 \
-upperf 4000 \
-wlen 0.025 \
-transform dct \
-round_filters no \
-remove_dc yes \
-svspec 0-12/13-25/26-38 \
-feat 1s_c_d_dd \
-agc none \
-cmn current \
-cmninit 47 \
-varnorm no

Current 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 47
-dither no no
-doublebw no no
-feat 1s_c_d_dd 1s_c_d_dd
-frate 100 100
-input_endian little little
-lda
-ldadim 0 0
-lifter 0 0
-logspec no no
-lowerf 133.33334 1.000000e+00
-ncep 13 13
-nfft 512 512
-nfilt 40 20
-remove_dc no yes
-round_filters yes no
-samprate 16000 1.600000e+04
-seed -1 -1
-smoothspec no no
-svspec 0-12/13-25/26-38
-transform legacy dct
-unit_area yes yes
-upperf 6855.4976 4.000000e+03
-varnorm no no
-verbose no no
-warp_params
-warp_type inverse_linear inverse_linear
-wlen 0.025625 2.500000e-02

INFO: file_omitted(0): Parsed model-specific feature parameters from /private/var/mobile/Containers/Bundle/Application/764368A1-D88E-4705-A908-103D2213B5D5/TimeFlowV2.app/AcousticModelEnglish.bundle/feat.params
INFO: file_omitted(0): Reading HMM transition probability matrices: /private/var/mobile/Containers/Bundle/Application/764368A1-D88E-4705-A908-103D2213B5D5/TimeFlowV2.app/AcousticModelEnglish.bundle/transition_matrices
INFO: file_omitted(0): Loading senones from dump file /private/var/mobile/Containers/Bundle/Application/764368A1-D88E-4705-A908-103D2213B5D5/TimeFlowV2.app/AcousticModelEnglish.bundle/sendump
INFO: file_omitted(0): BEGIN FILE FORMAT DESCRIPTION
INFO: file_omitted(0): Using memory-mapped I/O for senones
INFO: file_omitted(0): Building PID tables for dictionary
INFO: file_omitted(0): Allocating 50^3 * 2 bytes (244 KiB) for word-initial triphones
INFO: file_omitted(0): Allocated 30200 bytes (29 KiB) for word-final triphones
INFO: file_omitted(0): Allocated 30200 bytes (29 KiB) for single-phone word triphones
ERROR: “file_omitted”, line 0: File /var/mobile/Containers/Data/Application/7314BE0C-59C4-46C3-ACD2-13851947528E/Library/Caches/TimeFlowGrammerRuleSet.DMP not found
INFO: file_omitted(0): Number of word mappings: 100
INFO: file_omitted(0): Number of word mappings: 100
INFO: file_omitted(0): Initialized arc buffer ‘fwdtree’, each arc occupies 12 bytes
2014-11-20 15:23:32.349 TimeFlowV2[752:104246] Starting openAudioDevice on the device.
2014-11-20 15:23:32.350 TimeFlowV2[752:104246] Audio unit wrapper successfully created.
2014-11-20 15:23:32.363 TimeFlowV2[752:104246] Set audio route to SpeakerAndMicrophone
2014-11-20 15:23:32.365 TimeFlowV2[752:104246] Restoring SmartCMN value of 54.763176
2014-11-20 15:23:32.365 TimeFlowV2[752:104246] Checking and resetting all audio session settings.
2014-11-20 15:23:32.367 TimeFlowV2[752:104246] audioCategory is correct, we will leave it as it is.
2014-11-20 15:23:32.367 TimeFlowV2[752:104246] bluetoothInput is correct, we will leave it as it is.
2014-11-20 15:23:32.368 TimeFlowV2[752:104246] Output Device: SpeakerAndMicrophone.
2014-11-20 15:23:32.369 TimeFlowV2[752:104246] categoryDefaultToSpeaker is correct, we will leave it as it is.
2014-11-20 15:23:32.369 TimeFlowV2[752:104246] preferredBufferSize is incorrect, we will change it.
2014-11-20 15:23:32.371 TimeFlowV2[752:104246] PreferredBufferSize is now on the correct setting of 0.128000.
2014-11-20 15:23:32.373 TimeFlowV2[752:104246] preferredSampleRateCheck is correct, we will leave it as it is.
2014-11-20 15:23:32.375 TimeFlowV2[752:104246] Setting the variables for the device and starting it.
2014-11-20 15:23:32.377 TimeFlowV2[752:104246] Looping through ringbuffer sections and pre-allocating them.

2014-11-20 15:23:32.704 TimeFlowV2[752:104246] Started audio output unit.
2014-11-20 15:23:32.705 TimeFlowV2[752:104246] Calibration has started
2014-11-20 15:23:32.729 TimeFlowV2[752:104140] Audio route has changed for the following reason:
2014-11-20 15:23:32.730 TimeFlowV2[752:104140] There has been a change of category
2014-11-20 15:23:32.730 TimeFlowV2[752:104140] The previous audio route was Speaker
2014-11-20 15:23:32.731 TimeFlowV2[752:104140] This is not a case in which OpenEars performs a route change voluntarily. At the close of this function, the audio route is SpeakerAndMicrophone
2014-11-20 15:23:34.908 TimeFlowV2[752:104246] Calibration has completed
2014-11-20 15:23:34.910 TimeFlowV2[752:104246] Listening.
2014-11-20 15:23:38.659 TimeFlowV2[752:104246] Speech detected…
INFO: file_omitted(0): arc_buffer_producer_start_utt
time delta 1416518618.661699 start 0000000
INFO: file_omitted(0): arc_buffer_consumer_start_utt
time delta 1416518618.664940 start 0000000
INFO: file_omitted(0): cmn_prior_update: from < 54.76 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: file_omitted(0): cmn_prior_update: to < 46.89 -4.59 -1.65 -1.14 -2.20 -2.61 -0.43 -0.06 -0.71 -0.06 -0.58 -0.12 -0.25 >
INFO: file_omitted(0): idx 103 is final (103)
INFO: file_omitted(0): fwdtree: marking arc buffer final
INFO: file_omitted(0): fwdtree: allocated 32 arcs (0 KiB)
INFO: file_omitted(0): fwdtree: allocated 128 start frame entries (0 KiB)
INFO: file_omitted(0): Waiting for 1 consumers to finish
time delta 1416518620.341692 partial result fwdflat
partial: <s>:(0000000)
time delta 1416518620.349776 partial result fwdflat
partial: <s>:CREATE#^#TIMER:(0000000)
time delta 1416518620.352812 partial result fwdflat
partial: <s>:CREATE#^#TIMER:<sil>:(0000000)
INFO: file_omitted(0): idx 103 is final (103)
time delta 1416518620.353899 final result fwdflat
full: <s>:CREATE#^#TIMER:time delta 1416518620.354005 <sil>:end 0000000
2014-11-20 15:23:40.354 TimeFlowV2[752:104140] I’m running flite
2014-11-20 15:23:40.355 TimeFlowV2[752:104140] Checking and resetting all audio session settings.
2014-11-20 15:23:40.356 TimeFlowV2[752:104140] audioCategory is correct, we will leave it as it is.
2014-11-20 15:23:40.357 TimeFlowV2[752:104140] bluetoothInput is correct, we will leave it as it is.
2014-11-20 15:23:40.357 TimeFlowV2[752:104140] Output Device: SpeakerAndMicrophone.
2014-11-20 15:23:40.358 TimeFlowV2[752:104140] categoryDefaultToSpeaker is correct, we will leave it as it is.
2014-11-20 15:23:40.359 TimeFlowV2[752:104140] preferredBufferSize is incorrect, we will change it.
2014-11-20 15:23:40.360 TimeFlowV2[752:104140] PreferredBufferSize is now on the correct setting of 0.128000.
2014-11-20 15:23:40.361 TimeFlowV2[752:104140] preferredSampleRateCheck is correct, we will leave it as it is.
2014-11-20 15:23:40.734 TimeFlowV2[752:104140] I’m done running flite and it took 0.379356 seconds
2014-11-20 15:23:40.734 TimeFlowV2[752:104140] Flite audio player was nil when referenced so attempting to allocate a new audio player.
2014-11-20 15:23:40.735 TimeFlowV2[752:104140] Loading speech data for Flite concluded successfully.
(0000000)
time delta 1416518620.770111 end 0000000
2014-11-20 15:23:40.769 TimeFlowV2[752:104246] there is a request to change to the language model file /var/mobile/Containers/Data/Application/7314BE0C-59C4-46C3-ACD2-13851947528E/Library/Caches/TimeFlowGrammerRuleSet2.DMP. It is only possible to change language models with a valid ARPA file so if you experience a crash or error, not making a change language model request using an ARPA language model is the most likely reason.
2014-11-20 15:23:40.773 TimeFlowV2[752:104140] Flite sending suspend recognition notification.
2014-11-20 15:23:40.777 TimeFlowV2[752:104246] Stopping audio unit.
2014-11-20 15:23:40.898 TimeFlowV2[752:104246] Audio Output Unit stopped, cleaning up variable states.
2014-11-20 15:23:40.899 TimeFlowV2[752:104246] This device is not recording, so first we will set its recording status to 0
2014-11-20 15:23:40.900 TimeFlowV2[752:104246] The audio unit is running so we are going to dispose of its instance
INFO: file_omitted(0): fwdflat canceled
INFO: file_omitted(0): fwdtree canceled
INFO: file_omitted(0): TOTAL fwdtree 0.599 wall 0.582 xRT
INFO: file_omitted(0): TOTAL fwdflat 6.874 wall 6.674 xRT
INFO: file_omitted(0): Parsing command line:
\
-lm /var/mobile/Containers/Data/Application/7314BE0C-59C4-46C3-ACD2-13851947528E/Library/Caches/TimeFlowGrammerRuleSet2.DMP \
-beam 1e-66 \
-bestpath yes \
-dict /var/mobile/Containers/Data/Application/7314BE0C-59C4-46C3-ACD2-13851947528E/Library/Caches/TimeFlowGrammerRuleSet2.dic \
-hmm /private/var/mobile/Containers/Bundle/Application/764368A1-D88E-4705-A908-103D2213B5D5/TimeFlowV2.app/AcousticModelEnglish.bundle \
-lw 6.5 \
-samprate 16000

Current configuration:
[NAME] [DEFLT] [VALUE]
-adchdr 0 0
-adcin no no
-agc none none
-agcthresh 2.0 2.000000e+00
-align
-alpha 0.97 9.700000e-01
-arcdumpdir
-ascale 20.0 2.000000e+01
-aw 1 1
-backtrace no no
-beam 1e-48 1.000000e-66
-bestpath yes yes
-bestpathlw 9.5 9.500000e+00
-bghist no no
-cepdir
-cepext .mfc .mfc
-ceplen 13 13
-cmn current current
-cmninit 8.0 8.0
-compallsen no no
-ctl
-ctlcount -1 -1
-ctlincr 1 1
-ctloffset 0 0
-debug 0
-dict /var/mobile/Containers/Data/Application/7314BE0C-59C4-46C3-ACD2-13851947528E/Library/Caches/TimeFlowGrammerRuleSet2.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
-fwdflatlw 8.5 8.500000e+00
-fwdflatsfwin 25 25

-fwdflatwbeam 7e-29 7.000000e-29
-fwdtree yes yes
-fwdtreelm
-hmm /private/var/mobile/Containers/Bundle/Application/764368A1-D88E-4705-A908-103D2213B5D5/TimeFlowV2.app/AcousticModelEnglish.bundle
-hyp
-hypprefix
-input_endian little little
-jsgf
-kdmaxbbi -1 -1
-kdmaxdepth 0 0
-kdtree
-latsize 512 512
-lda
-ldadim 0 0
-lextreedump 0 0
-lifter 0 0
-lm /var/mobile/Containers/Data/Application/7314BE0C-59C4-46C3-ACD2-13851947528E/Library/Caches/TimeFlowGrammerRuleSet2.DMP
-lmctl
-lmname default default
-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 -1 -1
-maxnewoov 20 20
-maxsilfr -1 -1
-maxwpf 50 50
-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-5 1.000000e-05
-pl_window 0 0
-rawlogdir
-remove_dc no no
-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
-usewdphones no no
-uw 1.0 1.000000e+00
-var
-varfloor 0.0001 1.000000e-04
-varnorm no no
-verbose no no
-vm
-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-02

INFO: file_omitted(0): Parsing command line:
\
-nfilt 20 \
-lowerf 1 \
-upperf 4000 \
-wlen 0.025 \
-transform dct \
-round_filters no \
-remove_dc yes \
-svspec 0-12/13-25/26-38 \
-feat 1s_c_d_dd \
-agc none \
-cmn current \
-cmninit 47 \
-varnorm no

Current 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 47
-dither no no
-doublebw no no
-feat 1s_c_d_dd 1s_c_d_dd
-frate 100 100
-input_endian little little
-lda
-ldadim 0 0
-lifter 0 0
-logspec no no
-lowerf 133.33334 1.000000e+00
-ncep 13 13
-nfft 512 512
-nfilt 40 20
-remove_dc no yes
-round_filters yes no
-samprate 16000 1.600000e+04
-seed -1 -1
-smoothspec no no
-svspec 0-12/13-25/26-38
-transform legacy dct
-unit_area yes yes
-upperf 6855.4976 4.000000e+03
-varnorm no no
-verbose no no
-warp_params
-warp_type inverse_linear inverse_linear
-wlen 0.025625 2.500000e-02

INFO: file_omitted(0): Parsed model-specific feature parameters from /private/var/mobile/Containers/Bundle/Application/764368A1-D88E-4705-A908-103D2213B5D5/TimeFlowV2.app/AcousticModelEnglish.bundle/feat.params
INFO: file_omitted(0): Reading HMM transition probability matrices: /private/var/mobile/Containers/Bundle/Application/764368A1-D88E-4705-A908-103D2213B5D5/TimeFlowV2.app/AcousticModelEnglish.bundle/transition_matrices
INFO: file_omitted(0): Loading senones from dump file /private/var/mobile/Containers/Bundle/Application/764368A1-D88E-4705-A908-103D2213B5D5/TimeFlowV2.app/AcousticModelEnglish.bundle/sendump
INFO: file_omitted(0): BEGIN FILE FORMAT DESCRIPTION
INFO: file_omitted(0): Using memory-mapped I/O for senones
INFO: file_omitted(0): Building PID tables for dictionary
INFO: file_omitted(0): Allocating 50^3 * 2 bytes (244 KiB) for word-initial triphones
INFO: file_omitted(0): Allocated 30200 bytes (29 KiB) for word-final triphones
INFO: file_omitted(0): Allocated 30200 bytes (29 KiB) for single-phone word triphones
ERROR: “file_omitted”, line 0: File /var/mobile/Containers/Data/Application/7314BE0C-59C4-46C3-ACD2-13851947528E/Library/Caches/TimeFlowGrammerRuleSet2.DMP not found
INFO: file_omitted(0): Number of word mappings: 725
INFO: file_omitted(0): Number of word mappings: 725
INFO: file_omitted(0): Initialized arc buffer ‘fwdtree’, each arc occupies 12 bytes
2014-11-20 15:23:41.045 TimeFlowV2[752:104246] Starting openAudioDevice on the device.
2014-11-20 15:23:41.046 TimeFlowV2[752:104246] Audio unit wrapper successfully created.
2014-11-20 15:23:41.055 TimeFlowV2[752:104246] Set audio route to SpeakerAndMicrophone
2014-11-20 15:23:41.057 TimeFlowV2[752:104246] Restoring SmartCMN value of 46.893669
2014-11-20 15:23:41.057 TimeFlowV2[752:104246] Checking and resetting all audio session settings.
2014-11-20 15:23:41.059 TimeFlowV2[752:104246] audioCategory is correct, we will leave it as it is.
2014-11-20 15:23:41.060 TimeFlowV2[752:104246] bluetoothInput is correct, we will leave it as it is.
2014-11-20 15:23:41.061 TimeFlowV2[752:104246] Output Device: SpeakerAndMicrophone.
2014-11-20 15:23:41.061 TimeFlowV2[752:104246] categoryDefaultToSpeaker is correct, we will leave it as it is.
2014-11-20 15:23:41.062 TimeFlowV2[752:104246] preferredBufferSize is correct, we will leave it as it is.
2014-11-20 15:23:41.063 TimeFlowV2[752:104246] preferredSampleRateCheck is correct, we will leave it as it is.
2014-11-20 15:23:41.063 TimeFlowV2[752:104246] Setting the variables for the device and starting it.
2014-11-20 15:23:41.064 TimeFlowV2[752:104246] Looping through ringbuffer sections and pre-allocating them.
2014-11-20 15:23:41.067 TimeFlowV2[752:104246] Started audio output unit.
2014-11-20 15:23:41.068 TimeFlowV2[752:104246] Listening.
2014-11-20 15:23:41.715 TimeFlowV2[752:104246] Speech detected…
INFO: file_omitted(0): arc_buffer_producer_start_utt
time delta 1416518621.716527 start 0000001
INFO: file_omitted(0): arc_buffer_consumer_start_utt
time delta 1416518621.719415 start 0000001
2014-11-20 15:23:42.434 TimeFlowV2[752:104140] AVAudioPlayer did finish playing with success flag of 1
2014-11-20 15:23:42.586 TimeFlowV2[752:104140] Flite sending resume recognition notification.
INFO: file_omitted(0): cmn_prior_update: from < 46.89 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: file_omitted(0): cmn_prior_update: to < 53.92 -7.42 0.74 -3.86 -1.99 -0.30 -0.69 -0.51 -0.38 0.55 -0.75 0.42 0.12 >
INFO: file_omitted(0): idx 137 is final (137)
INFO: file_omitted(0): fwdtree: marking arc buffer final
INFO: file_omitted(0): fwdtree: allocated 256 arcs (3 KiB)
INFO: file_omitted(0): fwdtree: allocated 256 start frame entries (1 KiB)
INFO: file_omitted(0): Waiting for 1 consumers to finish
time delta 1416518623.873590 partial result fwdflat
partial: <s>:(0000001)
time delta 1416518623.880837 partial result fwdflat
partial: <s>:SET:(0000001)
time delta 1416518623.892117 partial result fwdflat
partial: <s>:SET#^#8#^#HOUR:(0000001)
time delta 1416518623.907454 partial result fwdflat
partial: <s>:SET#^#8#^#HOUR:8#^#HOUR:(0000001)
time delta 1416518623.911900 partial result fwdflat
partial: <s>:SET#^#8#^#HOUR:8#^#SECONDS:(0000001)
time delta 1416518623.922615 partial result fwdflat
partial: <s>:SET#^#8#^#HOUR:8#^#SECONDS:<sil>:(0000001)
INFO: file_omitted(0): idx 137 is final (137)
time delta 1416518623.925594 end 0000001
time delta 1416518623.925648 final result fwdflat
full: <s>:SET#^#8#^#HOUR:8#^#SECONDS:<sil>:(0000001)
time delta 1416518623.929309 end 0000001
2014-11-20 15:23:43.936 TimeFlowV2[752:104246] Stopping audio unit.
2014-11-20 15:23:43.969 TimeFlowV2[752:104246] Audio Output Unit stopped, cleaning up variable states.
2014-11-20 15:23:43.969 TimeFlowV2[752:104246] This device is not recording, so first we will set its recording status to 0
2014-11-20 15:23:43.970 TimeFlowV2[752:104246] The audio unit is running so we are going to dispose of its instance
INFO: file_omitted(0): fwdtree canceled
INFO: file_omitted(0): TOTAL fwdtree 0.216 wall 0.158 xRT
INFO: file_omitted(0): fwdflat canceled
INFO: file_omitted(0): TOTAL fwdflat 0.773 wall 0.564 xRT
2014-11-20 15:23:43.981 TimeFlowV2[752:104246] No longer listening.

[adb]

To clarify, all I am doing is checking for hypothesis data in rapidEarsDidReceiveFinishedSpeechHypothesis: and running flite like so [self.fliteController say:@”SET THE DURATION” withVoice:self.slt];. I do run a changeLanguageModelToFile right before this if that makes any difference. RapidEars picks up audio from flite and since I am listening for a command that begins with “SET” in my rule it thinks it hears set and returns a result.

[Halle Winkler]

Do you get a different outcome when you don’t change models right before?

[adb]

No, I commented it out and still had the same results.

[Halle Winkler]

Is it hearing the entire phrase or a subset?

[adb]

Sorry for the delayed response. It appears to be hearing all of it. I will change my rule to match exactly what flite is saying and test that quick.

[adb]

It is definitely hearing the whole phrase flite is saying. RapidEars matched the phrase exactly in rapidEarsDidReceiveFinishedSpeechHypothesis:.

[adb]

I also just verified that this happens even in the OpenEarsSampleApp + RapidEars. Place [self.fliteController say:@”GO FORWARD TURN LEFT” withVoice:self.slt]; in rapidEarsDidReceiveLiveSpeechHypothesis: and you create yourself a nice voice recognition infinite loop.

[Halle Winkler]

OK, I’ll take a look at it.

[adb]

Hey Halle,

Any possible update on this matter? Thanks for looking into it.

[Halle Winkler]

This is fixed in the next version which is going to be in testing for a bit – if you need a workaround now, just suspend in fliteDidStartSpeaking and resume in fliteDidFinishSpeaking, e.g.:

- (void) fliteDidStartSpeaking {
    [self.pocketsphinxController suspendRecognition];
}

- (void) fliteDidFinishSpeaking {
    [self.pocketsphinxController resumeRecognition];
}

I tested this in the sample app and it works as expected.

[adb]

Thanks for the update Halle. I appreciate the effort.

[Author]

Posts