Tagged: calibration, wake-up, word
- This topic has 42 replies, 2 voices, and was last updated 10 years, 2 months ago by Halle Winkler.
-
AuthorPosts
-
January 20, 2014 at 3:24 pm #1019807fdimParticipant
I am surprised that I haven’t found any similar topics. I want to create a wake-up word scenario in which I need openears to be continuously monitoring for the wake-up keyword “OK OPENEARS”. I use a language model that contains only the wake up keyword for that. When the wake-up keyword is spotted I change the Language Model to one that contains a variety of commands. And then I change back to the “wake-up” language model after command recognition success. Questions:
1) Should I use Rejecto or JSGF grammar for the wake-up language model?
2) In addition to the above I assume that wake-up keyword monitoring happens during music playback (from the device). When the keyword is spotted I change the language model as I previously said but I also turn the music down waiting for a command to be spoken. That means that the background noise level changes dramatically and OpenEars needs to recalibrate. Is there a fast way to force a predefined level of calibration?
3) Is this the best practice for my scenario?January 20, 2014 at 6:13 pm #1019810Halle WinklerPolitepixWelcome,
I am surprised that I haven’t found any similar topics.
I think it’s only that this is a specific version of a general implementation question that gets asked, i.e. how to spot a keyword. I’ll be happy to try to help you with it.
1) Should I use Rejecto or JSGF grammar for the wake-up language model?
You can’t use JSGF with a single keyword match, I think. JSGF rules as implemented in Pocketsphinx have to have more than one potential match or they crash, to the best of my knowledge.
Is there a fast way to force a predefined level of calibration?
You won’t get good results from a predefined level since it depends on the environment and mic characteristics being similar to your predefined case, which is unlikely. Have you already tested and discovered that the ongoing recalibration isn’t doing the job?
January 20, 2014 at 8:22 pm #1019811fdimParticipantActually here is what happens:
1) When I turn the music up OpenEars’ status is “Pocketsphinx has detected speech.” It never passes that phase as it thinks there is someone speaking instead of background music. Eventually it crashes. I need somehow to set a timeout for that and/or force a recalibration at this point.
2) When I start OpenEars with loud music (not insanely loud) from the beginning and before turning the music down I have this output:
2014-01-20 21:15:35.464 openEarsTest[6366:60b] Calibration time: 0
2014-01-20 21:15:36.163 openEarsTest[6366:60b] Pocketsphinx calibration has started.
2014-01-20 21:15:38.367 openEarsTest[6366:60b] Pocketsphinx calibration is complete.
2014-01-20 21:15:38.369 openEarsTest[6366:60b] Pocketsphinx is now listening.
2014-01-20 21:15:38.883 openEarsTest[6366:60b] Pocketsphinx has detected speech.
2014-01-20 21:15:47.429 openEarsTest[6366:60b] Pocketsphinx has detected a period of silence, concluding an utterance.
2014-01-20 21:15:47.743 openEarsTest[6366:60b] The received hypothesis is OK OK with a score of -998 and an ID of 000000000
2014-01-20 21:15:47.811 openEarsTest[6366:60b] Pocketsphinx is now listening.Then if I turn the music down it seems to get my first sentences wrong. However after a few seconds it starts getting them right again.
Any suggestions?
January 20, 2014 at 8:30 pm #1019812Halle WinklerPolitepixWhen I turn the music up OpenEars’ status is “Pocketsphinx has detected speech.” It never passes that phase as it thinks there is someone speaking instead of background music. Eventually it crashes. I need somehow to set a timeout for that and/or force a recalibration at this point.
Good to know — could you test against this beta that is designed to fix the issue where there is occasionally an endless recognition plus a crash when there is a sudden increase in volume in the background:
https://www.politepix.com/wp-content/uploads/OpenEarsDistributionBeta.tar.bz2
The beta is primarily designed for fixing this problem and so far I haven’t gotten any feedback on the beta from the other developers reporting the problem, so I’d appreciate hearing whether it alleviates the issue for you.
January 20, 2014 at 8:44 pm #1019813fdimParticipantI managed to crash it again with the beta. I am now trying to identify the exact steps that reproduce the problem.
January 20, 2014 at 8:49 pm #1019815fdimParticipantOK this is it:
2014-01-20 21:44:45.804 openEarsTest[6425:60b] Calibration time: 0
2014-01-20 21:44:46.482 openEarsTest[6425:60b] Pocketsphinx calibration has started.
2014-01-20 21:44:48.685 openEarsTest[6425:60b] Pocketsphinx calibration is complete.
2014-01-20 21:44:48.687 openEarsTest[6425:60b] Pocketsphinx is now listening.
2014-01-20 21:44:49.201 openEarsTest[6425:60b] Pocketsphinx has detected speech.
2014-01-20 21:44:52.374 openEarsTest[6425:60b] Pocketsphinx has detected a period of silence, concluding an utterance.
2014-01-20 21:44:52.558 openEarsTest[6425:60b] The received hypothesis is OK with a score of 0 and an ID of 000000000
2014-01-20 21:44:52.625 openEarsTest[6425:60b] Pocketsphinx is now listening.
2014-01-20 21:44:53.172 openEarsTest[6425:60b] Pocketsphinx has detected speech.After that there is a crash. I have turned up the music right before the last log message.
January 20, 2014 at 8:55 pm #1019816Halle WinklerPolitepixHi,
That’s the logging from your app — could you show me the output with verbosePocketsphinx and OpenEarsLogging turned on so I can get some info about what is happening with the framework?
Make sure you clean your project and rebuild with the new framework installed to make sure you’re linking to the new version.
Quick question: why does your log say that your calibration time is zero?
January 20, 2014 at 9:05 pm #1019817fdimParticipantNevermind that. That was a little line I added but I was printing before starting listening. If I print after beginning listening it returns 1.
Here is the detailed log. It kept listening about 3 minutes of Daft Punk before crashing. :)
2014-01-20 22:01:59.379 openEarsTest[6472:60b] Starting OpenEars logging for OpenEars version 1.64 on 32-bit device: iPad running iOS version: 7.000000
2014-01-20 22:01:59.387 openEarsTest[6472:60b] Normalized array contains the following entries:
(
“OK DEAR”
)
2014-01-20 22:01:59.433 openEarsTest[6472:60b] Starting dynamic language model generation
2014-01-20 22:01:59.438 openEarsTest[6472:60b] Able to open /var/mobile/Applications/79A5CC24-FD52-45CF-BFA2-FDF965D64F65/Library/Caches/NameIWantForMyLanguageModelFiles.corpus for reading
2014-01-20 22:01:59.440 openEarsTest[6472:60b] Able to open /var/mobile/Applications/79A5CC24-FD52-45CF-BFA2-FDF965D64F65/Library/Caches/NameIWantForMyLanguageModelFiles_pipe.txt for writing
2014-01-20 22:01:59.440 openEarsTest[6472:60b] Starting text2wfreq_impl
2014-01-20 22:01:59.456 openEarsTest[6472:60b] Done with text2wfreq_impl
2014-01-20 22:01:59.458 openEarsTest[6472:60b] Able to open /var/mobile/Applications/79A5CC24-FD52-45CF-BFA2-FDF965D64F65/Library/Caches/NameIWantForMyLanguageModelFiles_pipe.txt for reading.
2014-01-20 22:01:59.460 openEarsTest[6472:60b] Able to open /var/mobile/Applications/79A5CC24-FD52-45CF-BFA2-FDF965D64F65/Library/Caches/NameIWantForMyLanguageModelFiles.vocab for reading.
2014-01-20 22:01:59.461 openEarsTest[6472:60b] Starting wfreq2vocab
2014-01-20 22:01:59.464 openEarsTest[6472:60b] Done with wfreq2vocab
2014-01-20 22:01:59.466 openEarsTest[6472:60b] Starting text2idngram
2014-01-20 22:01:59.482 openEarsTest[6472:60b] Done with text2idngram
2014-01-20 22:01:59.486 openEarsTest[6472:60b] Starting idngram2lm2014-01-20 22:01:59.496 openEarsTest[6472:60b] Done with idngram2lm
2014-01-20 22:01:59.497 openEarsTest[6472:60b] Starting sphinx_lm_convert
2014-01-20 22:01:59.509 openEarsTest[6472:60b] Finishing sphinx_lm_convert
2014-01-20 22:01:59.514 openEarsTest[6472:60b] Done creating language model with CMUCLMTK in 0.079652 seconds.
2014-01-20 22:01:59.592 openEarsTest[6472:60b] I’m done running performDictionaryLookup and it took 0.048693 seconds
2014-01-20 22:01:59.598 openEarsTest[6472:60b] I’m done running dynamic language model generation and it took 411940919.598826 seconds
2014-01-20 22:01:59.601 openEarsTest[6472:60b] Leaving sample rate at the default of 16000.
2014-01-20 22:01:59.606 openEarsTest[6472:60b] The audio session has never been initialized so we will do that now.
2014-01-20 22:01:59.607 openEarsTest[6472:60b] Checking and resetting all audio session settings.
2014-01-20 22:01:59.609 openEarsTest[6472:60b] audioCategory is incorrect, we will change it.
2014-01-20 22:01:59.612 openEarsTest[6472:60b] audioCategory is now on the correct setting of kAudioSessionCategory_PlayAndRecord.
2014-01-20 22:01:59.614 openEarsTest[6472:60b] bluetoothInput is incorrect, we will change it.
2014-01-20 22:01:59.615 openEarsTest[6472:60b] bluetooth input is now on the correct setting of 1.
2014-01-20 22:01:59.617 openEarsTest[6472:60b] Output Device: SpeakerAndMicrophone.
2014-01-20 22:01:59.619 openEarsTest[6472:60b] categoryDefaultToSpeaker is incorrect, we will change it.
2014-01-20 22:01:59.620 openEarsTest[6472:60b] CategoryDefaultToSpeaker is now on the correct setting of 1.
2014-01-20 22:01:59.622 openEarsTest[6472:60b] preferredBufferSize is incorrect, we will change it.
2014-01-20 22:01:59.623 openEarsTest[6472:60b] PreferredBufferSize is now on the correct setting of 0.128000.
2014-01-20 22:01:59.624 openEarsTest[6472:60b] preferredSampleRateCheck is incorrect, we will change it.
2014-01-20 22:01:59.626 openEarsTest[6472:60b] preferred hardware sample rate is now on the correct setting of 16000.000000.
2014-01-20 22:01:59.659 openEarsTest[6472:60b] AudioSessionManager startAudioSession has reached the end of the initialization.
2014-01-20 22:01:59.660 openEarsTest[6472:60b] Exiting startAudioSession.
2014-01-20 22:01:59.663 openEarsTest[6472:3f07] Recognition loop has started
2014-01-20 22:01:59.834 openEarsTest[6472:3f07] Starting openAudioDevice on the device.
2014-01-20 22:01:59.835 openEarsTest[6472:3f07] Audio unit wrapper successfully created.
2014-01-20 22:01:59.844 openEarsTest[6472:3f07] Set audio route to SpeakerAndMicrophone
2014-01-20 22:01:59.846 openEarsTest[6472:3f07] Checking and resetting all audio session settings.
2014-01-20 22:01:59.848 openEarsTest[6472:3f07] audioCategory is correct, we will leave it as it is.
2014-01-20 22:01:59.849 openEarsTest[6472:3f07] bluetoothInput is correct, we will leave it as it is.
2014-01-20 22:01:59.851 openEarsTest[6472:3f07] Output Device: SpeakerAndMicrophone.
2014-01-20 22:01:59.852 openEarsTest[6472:3f07] categoryDefaultToSpeaker is correct, we will leave it as it is.
2014-01-20 22:01:59.854 openEarsTest[6472:3f07] preferredBufferSize is correct, we will leave it as it is.
2014-01-20 22:01:59.856 openEarsTest[6472:3f07] preferredSampleRateCheck is correct, we will leave it as it is.
2014-01-20 22:01:59.858 openEarsTest[6472:3f07] Setting the variables for the device and starting it.
2014-01-20 22:01:59.860 openEarsTest[6472:3f07] Looping through ringbuffer sections and pre-allocating them.
2014-01-20 22:02:00.330 openEarsTest[6472:3f07] Started audio output unit.
2014-01-20 22:02:00.333 openEarsTest[6472:3f07] Calibration has started
2014-01-20 22:02:00.334 openEarsTest[6472:60b] Pocketsphinx calibration has started.
2014-01-20 22:02:02.536 openEarsTest[6472:3f07] Calibration has completed
2014-01-20 22:02:02.537 openEarsTest[6472:60b] Pocketsphinx calibration is complete.
2014-01-20 22:02:02.538 openEarsTest[6472:3f07] Project has these words or phrases in its dictionary:
DEAR
OK
2014-01-20 22:02:02.540 openEarsTest[6472:3f07] Listening.
2014-01-20 22:02:02.541 openEarsTest[6472:60b] Pocketsphinx is now listening.
2014-01-20 22:02:03.711 openEarsTest[6472:3f07] Speech detected…
2014-01-20 22:02:03.712 openEarsTest[6472:60b] Pocketsphinx has detected speech.January 20, 2014 at 9:29 pm #1019818Halle WinklerPolitepixOK, can you also turn on verbosePocketsphinx? I have to see the logging from the recognition as well.
January 20, 2014 at 9:31 pm #1019819fdimParticipantI thought I did. This is the line I added:
pocketsphinxController.verbosePocketSphinx = YES;
in the begging of – (void)viewDidLoadJanuary 20, 2014 at 9:42 pm #1019838Halle WinklerPolitepixCan you put it inside the init code for PocketsphinxController instead, right after the init and alloc?
January 20, 2014 at 9:48 pm #1019839fdimParticipantOK it worked now:
2014-01-20 22:44:59.776 openEarsTest[6509:60b] Starting OpenEars logging for OpenEars version 1.64 on 32-bit device: iPad running iOS version: 7.000000
2014-01-20 22:44:59.783 openEarsTest[6509:60b] Normalized array contains the following entries:
(
“OK DEAR”
)
2014-01-20 22:44:59.825 openEarsTest[6509:60b] Starting dynamic language model generation
2014-01-20 22:44:59.830 openEarsTest[6509:60b] Able to open /var/mobile/Applications/79A5CC24-FD52-45CF-BFA2-FDF965D64F65/Library/Caches/NameIWantForMyLanguageModelFiles.corpus for reading
2014-01-20 22:44:59.832 openEarsTest[6509:60b] Able to open /var/mobile/Applications/79A5CC24-FD52-45CF-BFA2-FDF965D64F65/Library/Caches/NameIWantForMyLanguageModelFiles_pipe.txt for writing
2014-01-20 22:44:59.833 openEarsTest[6509:60b] Starting text2wfreq_impl
2014-01-20 22:44:59.851 openEarsTest[6509:60b] Done with text2wfreq_impl
2014-01-20 22:44:59.853 openEarsTest[6509:60b] Able to open /var/mobile/Applications/79A5CC24-FD52-45CF-BFA2-FDF965D64F65/Library/Caches/NameIWantForMyLanguageModelFiles_pipe.txt for reading.
2014-01-20 22:44:59.854 openEarsTest[6509:60b] Able to open /var/mobile/Applications/79A5CC24-FD52-45CF-BFA2-FDF965D64F65/Library/Caches/NameIWantForMyLanguageModelFiles.vocab for reading.
2014-01-20 22:44:59.856 openEarsTest[6509:60b] Starting wfreq2vocab
2014-01-20 22:44:59.858 openEarsTest[6509:60b] Done with wfreq2vocab
2014-01-20 22:44:59.860 openEarsTest[6509:60b] Starting text2idngram
2014-01-20 22:44:59.877 openEarsTest[6509:60b] Done with text2idngram
2014-01-20 22:44:59.881 openEarsTest[6509:60b] Starting idngram2lm2014-01-20 22:44:59.891 openEarsTest[6509:60b] Done with idngram2lm
2014-01-20 22:44:59.892 openEarsTest[6509:60b] Starting sphinx_lm_convert
2014-01-20 22:44:59.904 openEarsTest[6509:60b] Finishing sphinx_lm_convert
2014-01-20 22:44:59.909 openEarsTest[6509:60b] Done creating language model with CMUCLMTK in 0.082747 seconds.
2014-01-20 22:44:59.972 openEarsTest[6509:60b] I’m done running performDictionaryLookup and it took 0.050646 seconds
2014-01-20 22:44:59.978 openEarsTest[6509:60b] I’m done running dynamic language model generation and it took 411943499.978361 seconds
2014-01-20 22:44:59.980 openEarsTest[6509:60b] Leaving sample rate at the default of 16000.
2014-01-20 22:44:59.983 openEarsTest[6509:60b] The audio session has never been initialized so we will do that now.
2014-01-20 22:44:59.984 openEarsTest[6509:60b] Checking and resetting all audio session settings.
2014-01-20 22:44:59.986 openEarsTest[6509:60b] audioCategory is incorrect, we will change it.
2014-01-20 22:44:59.990 openEarsTest[6509:60b] audioCategory is now on the correct setting of kAudioSessionCategory_PlayAndRecord.
2014-01-20 22:44:59.992 openEarsTest[6509:60b] bluetoothInput is incorrect, we will change it.
2014-01-20 22:44:59.993 openEarsTest[6509:60b] bluetooth input is now on the correct setting of 1.
2014-01-20 22:44:59.995 openEarsTest[6509:60b] Output Device: SpeakerAndMicrophone.
2014-01-20 22:44:59.996 openEarsTest[6509:60b] categoryDefaultToSpeaker is incorrect, we will change it.
2014-01-20 22:44:59.998 openEarsTest[6509:60b] CategoryDefaultToSpeaker is now on the correct setting of 1.
2014-01-20 22:44:59.999 openEarsTest[6509:60b] preferredBufferSize is incorrect, we will change it.
2014-01-20 22:45:00.002 openEarsTest[6509:60b] PreferredBufferSize is now on the correct setting of 0.128000.
2014-01-20 22:45:00.005 openEarsTest[6509:60b] preferredSampleRateCheck is incorrect, we will change it.
2014-01-20 22:45:00.008 openEarsTest[6509:60b] preferred hardware sample rate is now on the correct setting of 16000.000000.
2014-01-20 22:45:00.045 openEarsTest[6509:60b] AudioSessionManager startAudioSession has reached the end of the initialization.
2014-01-20 22:45:00.046 openEarsTest[6509:60b] Exiting startAudioSession.
2014-01-20 22:45:00.050 openEarsTest[6509:4007] Recognition loop has started
INFO: file_omitted(0): Parsing command line:
\
-lm /var/mobile/Applications/79A5CC24-FD52-45CF-BFA2-FDF965D64F65/Library/Caches/NameIWantForMyLanguageModelFiles.DMP \
-dict /var/mobile/Applications/79A5CC24-FD52-45CF-BFA2-FDF965D64F65/Library/Caches/NameIWantForMyLanguageModelFiles.dic \
-hmm /var/mobile/Applications/79A5CC24-FD52-45CF-BFA2-FDF965D64F65/openEarsTest.app/AcousticModelEnglish.bundle \
-lw 6.500000 \
-samprate 16000Current configuration:
[NAME] [DEFLT] [VALUE]
-agc none none
-agcthresh 2.0 2.000000e+00
-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
-bghist no no
-ceplen 13 13
-cmn current current
-cmninit 8.0 8.0
-compallsen no no
-debug 0
-dict /var/mobile/Applications/79A5CC24-FD52-45CF-BFA2-FDF965D64F65/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 /var/mobile/Applications/79A5CC24-FD52-45CF-BFA2-FDF965D64F65/openEarsTest.app/AcousticModelEnglish.bundle
-input_endian little little
-jsgf
-kdmaxbbi -1 -1
-kdmaxdepth 0 0
-kdtree
-latsize 5000 5000
-lda
-ldadim 0 0
-lextreedump 0 0
-lifter 0 0
-lm /var/mobile/Applications/79A5CC24-FD52-45CF-BFA2-FDF965D64F65/Library/Caches/NameIWantForMyLanguageModelFiles.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
-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-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
-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 noCurrent 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-02INFO: file_omitted(0): Parsed model-specific feature parameters from /var/mobile/Applications/79A5CC24-FD52-45CF-BFA2-FDF965D64F65/openEarsTest.app/AcousticModelEnglish.bundle/feat.params
INFO: file_omitted(0): Initializing feature stream to type: ‘1s_c_d_dd’, ceplen=13, CMN=’current’, VARNORM=’no’, AGC=’none’
INFO: file_omitted(0): mean[0]= 12.00, mean[1..12]= 0.0
INFO: file_omitted(0): Using subvector specification 0-12/13-25/26-38
INFO: file_omitted(0): Reading model definition: /var/mobile/Applications/79A5CC24-FD52-45CF-BFA2-FDF965D64F65/openEarsTest.app/AcousticModelEnglish.bundle/mdef
INFO: file_omitted(0): Found byte-order mark BMDF, assuming this is a binary mdef file
INFO: file_omitted(0): Reading binary model definition: /var/mobile/Applications/79A5CC24-FD52-45CF-BFA2-FDF965D64F65/openEarsTest.app/AcousticModelEnglish.bundle/mdef
INFO: file_omitted(0): 50 CI-phone, 143047 CD-phone, 3 emitstate/phone, 150 CI-sen, 5150 Sen, 27135 Sen-Seq
INFO: file_omitted(0): Reading HMM transition probability matrices: /var/mobile/Applications/79A5CC24-FD52-45CF-BFA2-FDF965D64F65/openEarsTest.app/AcousticModelEnglish.bundle/transition_matrices
INFO: file_omitted(0): Attempting to use SCHMM computation module
INFO: file_omitted(0): Reading mixture gaussian parameter: /var/mobile/Applications/79A5CC24-FD52-45CF-BFA2-FDF965D64F65/openEarsTest.app/AcousticModelEnglish.bundle/means
INFO: file_omitted(0): 1 codebook, 3 feature, size:
INFO: file_omitted(0): 256×13
INFO: file_omitted(0): 256×13
INFO: file_omitted(0): 256×13
INFO: file_omitted(0): Reading mixture gaussian parameter: /var/mobile/Applications/79A5CC24-FD52-45CF-BFA2-FDF965D64F65/openEarsTest.app/AcousticModelEnglish.bundle/variances
INFO: file_omitted(0): 1 codebook, 3 feature, size:
INFO: file_omitted(0): 256×13
INFO: file_omitted(0): 256×13
INFO: file_omitted(0): 256×13
INFO: file_omitted(0): 0 variance values floored
INFO: file_omitted(0): Loading senones from dump file /var/mobile/Applications/79A5CC24-FD52-45CF-BFA2-FDF965D64F65/openEarsTest.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): Maximum top-N: 4 Top-N beams: 0 0 0
INFO: file_omitted(0): Allocating 4109 * 20 bytes (80 KiB) for word entries
INFO: file_omitted(0): Reading main dictionary: /var/mobile/Applications/79A5CC24-FD52-45CF-BFA2-FDF965D64F65/Library/Caches/NameIWantForMyLanguageModelFiles.dic
INFO: file_omitted(0): Allocated 0 KiB for strings, 0 KiB for phones
INFO: file_omitted(0): 2 words read
INFO: file_omitted(0): Reading filler dictionary: /var/mobile/Applications/79A5CC24-FD52-45CF-BFA2-FDF965D64F65/openEarsTest.app/AcousticModelEnglish.bundle/noisedict
INFO: file_omitted(0): Allocated 0 KiB for strings, 0 KiB for phones
INFO: file_omitted(0): 11 words read
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
INFO: file_omitted(0): No \data\ mark in LM file
INFO: file_omitted(0): Will use memory-mapped I/O for LM file
INFO: file_omitted(0): ngrams 1=4, 2=3, 3=2
INFO: file_omitted(0): 4 = LM.unigrams(+trailer) read
INFO: file_omitted(0): 3 = LM.bigrams(+trailer) read
INFO: file_omitted(0): 2 = LM.trigrams read
INFO: file_omitted(0): 2 = LM.prob2 entries read
INFO: file_omitted(0): 3 = LM.bo_wt2 entries read
INFO: file_omitted(0): 2 = LM.prob3 entries read
INFO: file_omitted(0): 1 = LM.tseg_base entries read
INFO: file_omitted(0): 4 = ascii word strings read
INFO: file_omitted(0): 2 unique initial diphones
INFO: file_omitted(0): 0 root, 0 non-root channels, 12 single-phone words
INFO: file_omitted(0): Creating search tree
INFO: file_omitted(0): before: 0 root, 0 non-root channels, 12 single-phone words
INFO: file_omitted(0): after: max nonroot chan increased to 130
INFO: file_omitted(0): after: 2 root, 2 non-root channels, 11 single-phone words
INFO: file_omitted(0): fwdflat: min_ef_width = 4, max_sf_win = 25
2014-01-20 22:45:00.240 openEarsTest[6509:4007] Starting openAudioDevice on the device.
2014-01-20 22:45:00.241 openEarsTest[6509:4007] Audio unit wrapper successfully created.
2014-01-20 22:45:00.250 openEarsTest[6509:4007] Set audio route to SpeakerAndMicrophone
2014-01-20 22:45:00.252 openEarsTest[6509:4007] Checking and resetting all audio session settings.
2014-01-20 22:45:00.255 openEarsTest[6509:4007] audioCategory is correct, we will leave it as it is.
2014-01-20 22:45:00.257 openEarsTest[6509:4007] bluetoothInput is correct, we will leave it as it is.
2014-01-20 22:45:00.259 openEarsTest[6509:4007] Output Device: SpeakerAndMicrophone.
2014-01-20 22:45:00.261 openEarsTest[6509:4007] categoryDefaultToSpeaker is correct, we will leave it as it is.
2014-01-20 22:45:00.263 openEarsTest[6509:4007] preferredBufferSize is correct, we will leave it as it is.
2014-01-20 22:45:00.265 openEarsTest[6509:4007] preferredSampleRateCheck is correct, we will leave it as it is.
2014-01-20 22:45:00.266 openEarsTest[6509:4007] Setting the variables for the device and starting it.
2014-01-20 22:45:00.268 openEarsTest[6509:4007] Looping through ringbuffer sections and pre-allocating them.
2014-01-20 22:45:00.743 openEarsTest[6509:4007] Started audio output unit.
2014-01-20 22:45:00.749 openEarsTest[6509:60b] Pocketsphinx calibration has started.
2014-01-20 22:45:00.750 openEarsTest[6509:4007] Calibration has started
2014-01-20 22:45:02.956 openEarsTest[6509:4007] Calibration has completed
2014-01-20 22:45:02.956 openEarsTest[6509:60b] Pocketsphinx calibration is complete.
2014-01-20 22:45:02.959 openEarsTest[6509:4007] Project has these words or phrases in its dictionary:
DEAR
OK
2014-01-20 22:45:02.960 openEarsTest[6509:4007] Listening.
2014-01-20 22:45:02.961 openEarsTest[6509:60b] Pocketsphinx is now listening.
2014-01-20 22:45:09.956 openEarsTest[6509:4007] Speech detected…
2014-01-20 22:45:09.958 openEarsTest[6509:60b] Pocketsphinx has detected speech.
2014-01-20 22:45:14.568 openEarsTest[6509:4007] Stopping audio unit.
2014-01-20 22:45:14.568 openEarsTest[6509:60b] Pocketsphinx has detected a period of silence, concluding an utterance.
2014-01-20 22:45:14.696 openEarsTest[6509:4007] Audio Output Unit stopped, cleaning up variable states.
2014-01-20 22:45:14.698 openEarsTest[6509:4007] Processing speech, please wait…
INFO: file_omitted(0): cmn_prior_update: from < 47.00 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: file_omitted(0): cmn_prior_update: to < 48.80 -2.72 -3.04 -3.18 -2.24 -2.12 -2.02 -1.14 -0.36 -0.53 0.77 -0.58 0.13 >
INFO: file_omitted(0): 3910 words recognized (10/fr)
INFO: file_omitted(0): 32048 senones evaluated (79/fr)
INFO: file_omitted(0): 15094 channels searched (37/fr), 802 1st, 13496 last
INFO: file_omitted(0): 4417 words for which last channels evaluated (10/fr)
INFO: file_omitted(0): 771 candidate words for entering last phone (1/fr)
INFO: file_omitted(0): fwdtree 0.32 CPU 0.079 xRT
INFO: file_omitted(0): fwdtree 4.81 wall 1.189 xRT
INFO: file_omitted(0): Utterance vocabulary contains 3 words
INFO: file_omitted(0): 747 words recognized (2/fr)
INFO: file_omitted(0): 25265 senones evaluated (62/fr)
INFO: file_omitted(0): 13773 channels searched (34/fr)
INFO: file_omitted(0): 938 words searched (2/fr)
INFO: file_omitted(0): 331 word transitions (0/fr)
INFO: file_omitted(0): fwdflat 0.12 CPU 0.029 xRT
INFO: file_omitted(0): fwdflat 0.12 wall 0.029 xRT
INFO: file_omitted(0): </s> not found in last frame, using <sil>.403 instead
INFO: file_omitted(0): lattice start node <s>.0 end node <sil>.390
INFO: file_omitted(0): Eliminated 0 nodes before end node
INFO: file_omitted(0): Lattice has 40 nodes, 35 links
INFO: file_omitted(0): Normalizer P(O) = alpha(<sil>:390:403) = -5419750
INFO: file_omitted(0): Joint P(O,S) = -5442689 P(S|O) = -22939
INFO: file_omitted(0): bestpath 0.00 CPU 0.000 xRT
INFO: file_omitted(0): bestpath 0.00 wall 0.000 xRT
2014-01-20 22:45:14.822 openEarsTest[6509:4007] Pocketsphinx heard “OK OK OK OK” with a score of (-22939) and an utterance ID of 000000000.
2014-01-20 22:45:14.824 openEarsTest[6509:60b] The received hypothesis is OK OK OK OK with a score of -22939 and an ID of 000000000
2014-01-20 22:45:14.824 openEarsTest[6509:4007] Checking and resetting all audio session settings.
2014-01-20 22:45:14.827 openEarsTest[6509:4007] audioCategory is correct, we will leave it as it is.
2014-01-20 22:45:14.828 openEarsTest[6509:4007] bluetoothInput is correct, we will leave it as it is.
2014-01-20 22:45:14.829 openEarsTest[6509:4007] Output Device: SpeakerAndMicrophone.
2014-01-20 22:45:14.831 openEarsTest[6509:4007] categoryDefaultToSpeaker is correct, we will leave it as it is.
2014-01-20 22:45:14.832 openEarsTest[6509:4007] preferredBufferSize is correct, we will leave it as it is.
2014-01-20 22:45:14.833 openEarsTest[6509:4007] preferredSampleRateCheck is correct, we will leave it as it is.
2014-01-20 22:45:14.835 openEarsTest[6509:4007] Setting the variables for the device and starting it.
2014-01-20 22:45:14.838 openEarsTest[6509:4007] Looping through ringbuffer sections and pre-allocating them.
2014-01-20 22:45:14.948 openEarsTest[6509:4007] Started audio output unit.
2014-01-20 22:45:14.950 openEarsTest[6509:4007] Listening.
2014-01-20 22:45:14.952 openEarsTest[6509:60b] Pocketsphinx is now listening.
2014-01-20 22:45:16.430 openEarsTest[6509:4007] Speech detected…
2014-01-20 22:45:16.432 openEarsTest[6509:60b] Pocketsphinx has detected speech.
INFO: file_omitted(0): Resized backpointer table to 10000 entries
INFO: file_omitted(0): Resized backpointer table to 20000 entries
2014-01-20 22:45:33.127 openEarsTest[6509:4007] Stopping audio unit.
2014-01-20 22:45:33.127 openEarsTest[6509:60b] Pocketsphinx has detected a period of silence, concluding an utterance.
2014-01-20 22:45:33.255 openEarsTest[6509:4007] Audio Output Unit stopped, cleaning up variable states.
2014-01-20 22:45:33.256 openEarsTest[6509:4007] Processing speech, please wait…
INFO: file_omitted(0): cmn_prior_update: from < 49.83 -3.93 -4.26 -3.81 -2.32 -2.17 -1.42 -1.14 0.50 -0.20 0.53 -0.22 0.18 >
INFO: file_omitted(0): cmn_prior_update: to < 50.07 -3.98 -4.51 -3.67 -2.59 -2.27 -1.40 -0.98 0.65 -0.09 0.54 -0.19 0.23 >
INFO: file_omitted(0): 16489 words recognized (10/fr)
INFO: file_omitted(0): 142092 senones evaluated (89/fr)
INFO: file_omitted(0): 71282 channels searched (44/fr), 3196 1st, 64896 last
INFO: file_omitted(0): 17869 words for which last channels evaluated (11/fr)
INFO: file_omitted(0): 3079 candidate words for entering last phone (1/fr)
INFO: file_omitted(0): fwdtree 1.15 CPU 0.072 xRT
INFO: file_omitted(0): fwdtree 16.85 wall 1.052 xRT
INFO: file_omitted(0): Utterance vocabulary contains 3 words
INFO: file_omitted(0): 3179 words recognized (2/fr)
INFO: file_omitted(0): 134141 senones evaluated (84/fr)
INFO: file_omitted(0): 81140 channels searched (50/fr)
INFO: file_omitted(0): 4370 words searched (2/fr)
INFO: file_omitted(0): 2106 word transitions (1/fr)
INFO: file_omitted(0): fwdflat 0.46 CPU 0.029 xRT
INFO: file_omitted(0): fwdflat 0.47 wall 0.029 xRT
INFO: file_omitted(0): </s> not found in last frame, using DEAR.1600 instead
INFO: file_omitted(0): lattice start node <s>.0 end node DEAR.1584
INFO: file_omitted(0): Eliminated 5 nodes before end node
INFO: file_omitted(0): Lattice has 225 nodes, 153 links
INFO: file_omitted(0): Normalizer P(O) = alpha(DEAR:1584:1600) = -11540084
INFO: file_omitted(0): Joint P(O,S) = -11644886 P(S|O) = -104802
INFO: file_omitted(0): bestpath 0.01 CPU 0.001 xRT
INFO: file_omitted(0): bestpath 0.02 wall 0.001 xRT
2014-01-20 22:45:33.748 openEarsTest[6509:4007] Pocketsphinx heard “OK OK OK DEAR OK OK OK DEAR OK OK OK OK OK DEAR” with a score of (-104802) and an utterance ID of 000000001.
2014-01-20 22:45:33.749 openEarsTest[6509:4007] Checking and resetting all audio session settings.
2014-01-20 22:45:33.750 openEarsTest[6509:60b] The received hypothesis is OK OK OK DEAR OK OK OK DEAR OK OK OK OK OK DEAR with a score of -104802 and an ID of 000000001
2014-01-20 22:45:33.752 openEarsTest[6509:4007] audioCategory is correct, we will leave it as it is.
2014-01-20 22:45:33.753 openEarsTest[6509:4007] bluetoothInput is correct, we will leave it as it is.
2014-01-20 22:45:33.755 openEarsTest[6509:4007] Output Device: SpeakerAndMicrophone.
2014-01-20 22:45:33.756 openEarsTest[6509:4007] categoryDefaultToSpeaker is correct, we will leave it as it is.
2014-01-20 22:45:33.758 openEarsTest[6509:4007] preferredBufferSize is correct, we will leave it as it is.
2014-01-20 22:45:33.759 openEarsTest[6509:4007] preferredSampleRateCheck is correct, we will leave it as it is.
2014-01-20 22:45:33.760 openEarsTest[6509:4007] Setting the variables for the device and starting it.
2014-01-20 22:45:33.761 openEarsTest[6509:4007] Looping through ringbuffer sections and pre-allocating them.
2014-01-20 22:45:33.892 openEarsTest[6509:4007] Started audio output unit.
2014-01-20 22:45:33.894 openEarsTest[6509:4007] Listening.
2014-01-20 22:45:33.896 openEarsTest[6509:60b] Pocketsphinx is now listening.
2014-01-20 22:45:35.249 openEarsTest[6509:4007] Speech detected…
2014-01-20 22:45:35.251 openEarsTest[6509:60b] Pocketsphinx has detected speech.
INFO: file_omitted(0): Resized backpointer table to 40000 entries
INFO: file_omitted(0): Resized backpointer table to 80000 entries
INFO: file_omitted(0): Resized backpointer table to 160000 entries
INFO: file_omitted(0): Resized score stack to 200000 entries
INFO: file_omitted(0): Resized backpointer table to 320000 entries
FATAL_ERROR: “file_omitted”, line 0: Decoder can not process more than 32767 frames at once, requested 32768January 20, 2014 at 9:49 pm #1019840Halle WinklerPolitepixI’ve just updated the beta with the latest version — can you re-download and test against it instead? You’ll know it’s the right version because this line:
Starting OpenEars logging for OpenEars version 1.64
will say 1.65 instead. Thanks!
January 20, 2014 at 9:57 pm #1019844Halle WinklerPolitepix(I think it should fix the “Resized backpointer table” thing you can see at the end of your verbose logs, which is the root of the crash).
January 20, 2014 at 10:03 pm #1019845fdimParticipantI might be doing something wrong. I am still getting “Starting OpenEars logging for OpenEars version 1.64”. For Verification:
1) I download from this link http://cdn5.politepix.com/wp-content/uploads/OpenEarsDistributionBeta.tar.bz2
2) I delete references from the project on the old “framework” folder
3) Sent the folder to trash
4) Readd the new “framework” folder
5) Clean
6) BuildJanuary 20, 2014 at 10:13 pm #1019846Halle WinklerPolitepixWhoops, I’m sorry, that was my fault. I uploaded the project but I hadn’t built the framework first so it had the new source but the old framework. The current version I just uploaded should not have this problem, so you should be able to re-download and follow the same process you did before.
January 20, 2014 at 10:22 pm #1019847fdimParticipantSorry 1.64 again :(
Are you sure your are uploading to the correct path?January 20, 2014 at 10:43 pm #1019848Halle WinklerPolitepixYup, but that time the CDN didn’t really purge. OK, I’ve downloaded it and installed and run it to make sure it’s OK, so I promise if you download it this time you’ll get the right framework :) .
January 21, 2014 at 9:38 am #1019852fdimParticipantOK just checked. There are no crashes with this version. I will now do some extensive testing with my scenario and get back to you if anything else comes up.
January 21, 2014 at 9:49 am #1019853Halle WinklerPolitepixExcellent, thank you for testing.
January 22, 2014 at 9:19 am #1019861fdimParticipantI ran some tests and found out the following:
1) If I turn the music up while openears is already listening it goes to “Pocketsphinx has detected speech” status, after a while it says it detected silence and and goes back to “Pocketsphinx has detected speech” and all this happens recurrently. The obvious problem is that it never calibrates to the new noise level and of course it is not able to hear me.
2) On the contrary if I start openears during loud music playback it calibrates and works just fine. It is able to hear me if I speak over the music.
I think we need a method to force recalibration on demand.
January 22, 2014 at 9:36 am #1019862Halle WinklerPolitepixOK, I don’t think there is going to be an API change here (the design goal is for it to do the right thing instead of asking developers to micromanage the state), but I can test today with telling it to recalibrate under the circumstance that it returns from a sudden jump in noise level and then give you a beta to test.
January 22, 2014 at 9:40 am #1019863fdimParticipantThat would be nice. I might also add that the above would be handy even when a sudden drop in noise occurs as well.
January 22, 2014 at 9:44 am #1019864Halle WinklerPolitepixAre you very confident that your test of calibrating at the beginning versus raising the noise level later is an apples-to-apples test, i.e. with the exact same music, segment of that music, and volumes? The reason I ask is that the voice audio detection is already reset under the circumstances you’re describing, i.e. it is immediately taking in new noise floor info any time that it returns right after a big jump in noise.
It’s also a surprising result that loud music playing back at or above speech levels would not be categorized as non-silence whether it happened at the beginning of a new session or in the middle of one. It would normally be a correct categorization to call that non-silence because it is occurring at or above the volume level of speech and the voice activity detector doesn’t have a way of distinguishing between music-like sound and speech-like sound. If you are confident that you get a different result with an initial calibration against the exact same noise level I will be up for testing it and trying to improve that.
January 22, 2014 at 9:54 am #1019866fdimParticipantYes I am very confident. I did it as you said. Same track, same segment, same volume. I am afraid that the auto detection you describe does not work.
I agree with you and it is logical to confuse music with speech if they are at the same level. However at some point and after a few failed attempts to recognise that it is noise it is hearing and not speech, it should recalibrate and snap out of it. Right now it stays in the “detecting speech – detected silence” loop and practically it becomes unusable.
I am always referring to your last beta (just in case).
January 22, 2014 at 10:09 am #1019867Halle WinklerPolitepixOK, thank you for elaborating and I’ll check it out today.
January 24, 2014 at 11:50 am #1019894fdimParticipantAny news on this?
January 24, 2014 at 11:52 am #1019895Halle WinklerPolitepixNope, I will post when I have a new beta for testing.
January 28, 2014 at 7:17 pm #1019956Halle WinklerPolitepixOK, there is a new beta to test at:
https://www.politepix.com/wp-content/uploads/OpenEarsDistributionBeta.tar.bz2
This will recalibrate any time that it experiences a fast background noise level jump that could previously result in the VAD getting confused and hanging. It will use all the available samples in the ringbuffer up to that point in time to calibrate in place in a fraction of a second.
Loud noise levels will still be a challenging situation for recognition so it shouldn’t be expected that this will remove all traces of long or confused recognitions in environments with competing noise that is as loud or louder than speech — what we’re looking for here is that the VAD doesn’t become stuck in a particular state, but the usual compromised recognition results in environments with lots of noise and sudden noise level changes are to be expected.
I am also trying out an isSuspended and an isListening property for PocketsphinxController which you can check out in this version, although I am not committing to adding them for 1.65.
In order to add this calibration change and the other fixes while keeping things maintainable it was necessary to refactor the listening loop, so please let me know about any other odd behavior you see.
January 30, 2014 at 6:22 pm #1019993Halle WinklerPolitepixOK, this version will ship tomorrow along with the related plugin updates unless I hear from anyone that they are seeing issues with it.
January 30, 2014 at 8:16 pm #1019994fdimParticipantSorry for taking so long to reply. It has been a busy week.
I just tested it and I can definitely say that there has been an improvement. I have noticed some funny things that may or may not be an issue.
a) Everytime it recognises my command, right before it throws a detected silence message. Like this:
2014-01-30 21:10:33.968 openEarsTest[1636:60b] Pocketsphinx is now listening.
2014-01-30 21:10:35.304 openEarsTest[1636:60b] Pocketsphinx has detected speech.
2014-01-30 21:10:37.172 openEarsTest[1636:60b] Pocketsphinx has detected a period of silence, concluding an utterance.
2014-01-30 21:10:37.332 openEarsTest[1636:60b] The received hypothesis is OKDEAR with a score of -534 and an ID of 000000036b) I have noticed that every time the background noise changed (music on/off) my first three commands went to waste and after that it started recognising. This happens even when I turn the music off so there in no background noise.
I also had a crash after two minutes of inactivity.
I am trying to generate a detailed log that reflects all these issues.
January 30, 2014 at 8:33 pm #1019995fdimParticipantOK the crash is verified I think:
2014-01-30 21:26:25.776 openEarsTest[1700:60b] Starting OpenEars logging for OpenEars version 1.65 on 32-bit device: iPad running iOS version: 7.000000
2014-01-30 21:26:25.783 openEarsTest[1700:60b] Normalized array contains the following entries:
(
OKDEAR
)
2014-01-30 21:26:25.825 openEarsTest[1700:60b] Starting dynamic language model generation
2014-01-30 21:26:25.830 openEarsTest[1700:60b] Able to open /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/Library/Caches/NameIWantForMyLanguageModelFiles.corpus for reading
2014-01-30 21:26:25.833 openEarsTest[1700:60b] Able to open /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/Library/Caches/NameIWantForMyLanguageModelFiles_pipe.txt for writing
2014-01-30 21:26:25.834 openEarsTest[1700:60b] Starting text2wfreq_impl
2014-01-30 21:26:25.847 openEarsTest[1700:60b] Done with text2wfreq_impl
2014-01-30 21:26:25.849 openEarsTest[1700:60b] Able to open /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/Library/Caches/NameIWantForMyLanguageModelFiles_pipe.txt for reading.
2014-01-30 21:26:25.850 openEarsTest[1700:60b] Able to open /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/Library/Caches/NameIWantForMyLanguageModelFiles.vocab for reading.
2014-01-30 21:26:25.852 openEarsTest[1700:60b] Starting wfreq2vocab
2014-01-30 21:26:25.854 openEarsTest[1700:60b] Done with wfreq2vocab
2014-01-30 21:26:25.856 openEarsTest[1700:60b] Starting text2idngram
2014-01-30 21:26:25.872 openEarsTest[1700:60b] Done with text2idngram
2014-01-30 21:26:25.876 openEarsTest[1700:60b] Starting idngram2lm2014-01-30 21:26:25.885 openEarsTest[1700:60b] Done with idngram2lm
2014-01-30 21:26:25.886 openEarsTest[1700:60b] Starting sphinx_lm_convert
2014-01-30 21:26:25.896 openEarsTest[1700:60b] Finishing sphinx_lm_convert
2014-01-30 21:26:25.901 openEarsTest[1700:60b] Done creating language model with CMUCLMTK in 0.075254 seconds.
2014-01-30 21:26:25.991 openEarsTest[1700:60b] The word OKDEAR was not found in the dictionary /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/openEarsTest.app/AcousticModelEnglish.bundle/LanguageModelGeneratorLookupList.text/LanguageModelGeneratorLookupList.text.
2014-01-30 21:26:25.992 openEarsTest[1700:60b] Now using the fallback method to look up the word OKDEAR
2014-01-30 21:26:25.994 openEarsTest[1700:60b] 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, or that you are submitting the words in lowercase when they need to be entirely written in uppercase.
2014-01-30 21:26:25.999 openEarsTest[1700:60b] Using convertGraphemes for the word or phrase OKDEAR which doesn’t appear in the dictionary
2014-01-30 21:26:26.025 openEarsTest[1700:60b] I’m done running performDictionaryLookup and it took 0.112022 seconds
2014-01-30 21:26:26.030 openEarsTest[1700:60b] I’m done running dynamic language model generation and it took 0.248456 seconds
2014-01-30 21:26:26.033 openEarsTest[1700:60b] Leaving sample rate at the default of 16000.
2014-01-30 21:26:26.036 openEarsTest[1700:60b] The audio session has never been initialized so we will do that now.
2014-01-30 21:26:26.037 openEarsTest[1700:60b] Checking and resetting all audio session settings.
2014-01-30 21:26:26.038 openEarsTest[1700:60b] audioCategory is incorrect, we will change it.
2014-01-30 21:26:26.042 openEarsTest[1700:60b] audioCategory is now on the correct setting of kAudioSessionCategory_PlayAndRecord.
2014-01-30 21:26:26.043 openEarsTest[1700:60b] bluetoothInput is incorrect, we will change it.
2014-01-30 21:26:26.044 openEarsTest[1700:60b] bluetooth input is now on the correct setting of 1.
2014-01-30 21:26:26.046 openEarsTest[1700:60b] Output Device: SpeakerAndMicrophone.
2014-01-30 21:26:26.047 openEarsTest[1700:60b] categoryDefaultToSpeaker is incorrect, we will change it.
2014-01-30 21:26:26.049 openEarsTest[1700:60b] CategoryDefaultToSpeaker is now on the correct setting of 1.
2014-01-30 21:26:26.050 openEarsTest[1700:60b] preferredBufferSize is incorrect, we will change it.
2014-01-30 21:26:26.051 openEarsTest[1700:60b] PreferredBufferSize is now on the correct setting of 0.128000.
2014-01-30 21:26:26.052 openEarsTest[1700:60b] preferredSampleRateCheck is incorrect, we will change it.
2014-01-30 21:26:26.054 openEarsTest[1700:60b] preferred hardware sample rate is now on the correct setting of 16000.000000.
2014-01-30 21:26:26.079 openEarsTest[1700:60b] AudioSessionManager startAudioSession has reached the end of the initialization.
2014-01-30 21:26:26.082 openEarsTest[1700:60b] Exiting startAudioSession.
2014-01-30 21:26:26.088 openEarsTest[1700:4107] setSecondsOfSilence value of 0.000000 was too large or too small or was NULL, using default of 0.700000.
2014-01-30 21:26:26.090 openEarsTest[1700:4107] Project has these words or phrases in its dictionary:
OKDEAR
2014-01-30 21:26:26.092 openEarsTest[1700:4107] Recognition loop has started
INFO: file_omitted(0): Parsing command line:
\
-lm /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/Library/Caches/NameIWantForMyLanguageModelFiles.DMP \
-dict /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/Library/Caches/NameIWantForMyLanguageModelFiles.dic \
-hmm /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/openEarsTest.app/AcousticModelEnglish.bundle \
-lw 6.500000 \
-samprate 16000Current configuration:
[NAME] [DEFLT] [VALUE]
-agc none none
-agcthresh 2.0 2.000000e+00
-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
-bghist no no
-ceplen 13 13
-cmn current current
-cmninit 8.0 8.0
-compallsen no no
-debug 0
-dict /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/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 /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/openEarsTest.app/AcousticModelEnglish.bundle
-input_endian little little
-jsgf
-kdmaxbbi -1 -1
-kdmaxdepth 0 0
-kdtree
-latsize 5000 5000
-lda
-ldadim 0 0
-lextreedump 0 0
-lifter 0 0
-lm /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/Library/Caches/NameIWantForMyLanguageModelFiles.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
-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-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
-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-022014-01-30 21:26:26.105 openEarsTest[1700:60b] Audio route has changed for the following reason:
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 noCurrent 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-02INFO: file_omitted(0): Parsed model-specific feature parameters from /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/openEarsTest.app/AcousticModelEnglish.bundle/feat.params
2014-01-30 21:26:26.123 openEarsTest[1700:60b] There has been a change of category
INFO: file_omitted(0): Initializing feature stream to type: ‘1s_c_d_dd’, ceplen=13, CMN=’current’, VARNORM=’no’, AGC=’none’
INFO: file_omitted(0): mean[0]= 12.00, mean[1..12]= 0.0
INFO: file_omitted(0): Using subvector specification 0-12/13-25/26-38
INFO: file_omitted(0): Reading model definition: /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/openEarsTest.app/AcousticModelEnglish.bundle/mdef
INFO: file_omitted(0): Found byte-order mark BMDF, assuming this is a binary mdef file
INFO: file_omitted(0): Reading binary model definition: /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/openEarsTest.app/AcousticModelEnglish.bundle/mdef
2014-01-30 21:26:26.130 openEarsTest[1700:60b] The previous audio route was Speaker
2014-01-30 21:26:26.132 openEarsTest[1700:60b] This is not a case in which OpenEars performs a route change voluntarily. At the close of this function, the audio route is SpeakerAndMicrophone
INFO: file_omitted(0): 50 CI-phone, 143047 CD-phone, 3 emitstate/phone, 150 CI-sen, 5150 Sen, 27135 Sen-Seq
INFO: file_omitted(0): Reading HMM transition probability matrices: /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/openEarsTest.app/AcousticModelEnglish.bundle/transition_matrices
INFO: file_omitted(0): Attempting to use SCHMM computation module
INFO: file_omitted(0): Reading mixture gaussian parameter: /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/openEarsTest.app/AcousticModelEnglish.bundle/means
INFO: file_omitted(0): 1 codebook, 3 feature, size:
INFO: file_omitted(0): 256×13
INFO: file_omitted(0): 256×13
INFO: file_omitted(0): 256×13
INFO: file_omitted(0): Reading mixture gaussian parameter: /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/openEarsTest.app/AcousticModelEnglish.bundle/variances
INFO: file_omitted(0): 1 codebook, 3 feature, size:
INFO: file_omitted(0): 256×13
INFO: file_omitted(0): 256×13
INFO: file_omitted(0): 256×13
INFO: file_omitted(0): 0 variance values floored
INFO: file_omitted(0): Loading senones from dump file /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/openEarsTest.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): Maximum top-N: 4 Top-N beams: 0 0 0
INFO: file_omitted(0): Allocating 4108 * 20 bytes (80 KiB) for word entries
INFO: file_omitted(0): Reading main dictionary: /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/Library/Caches/NameIWantForMyLanguageModelFiles.dic
INFO: file_omitted(0): Allocated 0 KiB for strings, 0 KiB for phones
INFO: file_omitted(0): 1 words read
INFO: file_omitted(0): Reading filler dictionary: /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/openEarsTest.app/AcousticModelEnglish.bundle/noisedict
INFO: file_omitted(0): Allocated 0 KiB for strings, 0 KiB for phones
INFO: file_omitted(0): 11 words read
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
INFO: file_omitted(0): No \data\ mark in LM file
INFO: file_omitted(0): Will use memory-mapped I/O for LM file
INFO: file_omitted(0): ngrams 1=3, 2=2, 3=1
INFO: file_omitted(0): 3 = LM.unigrams(+trailer) read
INFO: file_omitted(0): 2 = LM.bigrams(+trailer) read
INFO: file_omitted(0): 1 = LM.trigrams read
INFO: file_omitted(0): 2 = LM.prob2 entries read
INFO: file_omitted(0): 3 = LM.bo_wt2 entries read
INFO: file_omitted(0): 2 = LM.prob3 entries read
INFO: file_omitted(0): 1 = LM.tseg_base entries read
INFO: file_omitted(0): 3 = ascii word strings read
INFO: file_omitted(0): 1 unique initial diphones
INFO: file_omitted(0): 0 root, 0 non-root channels, 12 single-phone words
INFO: file_omitted(0): Creating search tree
INFO: file_omitted(0): before: 0 root, 0 non-root channels, 12 single-phone words
INFO: file_omitted(0): after: max nonroot chan increased to 131
INFO: file_omitted(0): after: 1 root, 3 non-root channels, 11 single-phone words
INFO: file_omitted(0): fwdflat: min_ef_width = 4, max_sf_win = 25
2014-01-30 21:26:26.248 openEarsTest[1700:4107] Starting openAudioDevice on the device.
2014-01-30 21:26:26.249 openEarsTest[1700:4107] Audio unit wrapper successfully created.
2014-01-30 21:26:26.256 openEarsTest[1700:4107] Set audio route to SpeakerAndMicrophone
2014-01-30 21:26:26.257 openEarsTest[1700:4107] There is no CMN plist so we are using the fresh CMN value 47.000000.
2014-01-30 21:26:26.258 openEarsTest[1700:4107] Checking and resetting all audio session settings.
2014-01-30 21:26:26.259 openEarsTest[1700:4107] audioCategory is correct, we will leave it as it is.
2014-01-30 21:26:26.261 openEarsTest[1700:4107] bluetoothInput is correct, we will leave it as it is.
2014-01-30 21:26:26.262 openEarsTest[1700:4107] Output Device: SpeakerAndMicrophone.
2014-01-30 21:26:26.263 openEarsTest[1700:4107] categoryDefaultToSpeaker is correct, we will leave it as it is.
2014-01-30 21:26:26.264 openEarsTest[1700:4107] preferredBufferSize is correct, we will leave it as it is.
2014-01-30 21:26:26.265 openEarsTest[1700:4107] preferredSampleRateCheck is correct, we will leave it as it is.
2014-01-30 21:26:26.266 openEarsTest[1700:4107] Setting the variables for the device and starting it.
2014-01-30 21:26:26.267 openEarsTest[1700:4107] Looping through ringbuffer sections and pre-allocating them.
2014-01-30 21:26:26.743 openEarsTest[1700:4107] Started audio output unit.
2014-01-30 21:26:26.746 openEarsTest[1700:4107] Calibration has started
2014-01-30 21:26:26.746 openEarsTest[1700:60b] Pocketsphinx calibration has started.
2014-01-30 21:26:28.950 openEarsTest[1700:4107] Calibration has completed
2014-01-30 21:26:28.951 openEarsTest[1700:60b] Pocketsphinx calibration is complete.
2014-01-30 21:26:28.954 openEarsTest[1700:4107] Listening.
2014-01-30 21:26:28.955 openEarsTest[1700:60b] Pocketsphinx is now listening.
2014-01-30 21:26:31.533 openEarsTest[1700:4107] Speech detected…
2014-01-30 21:26:31.535 openEarsTest[1700:60b] Pocketsphinx has detected speech.
2014-01-30 21:26:33.273 openEarsTest[1700:4107] Stopping audio unit.
2014-01-30 21:26:33.274 openEarsTest[1700:60b] Pocketsphinx has detected a period of silence, concluding an utterance.
2014-01-30 21:26:33.402 openEarsTest[1700:4107] Audio Output Unit stopped, cleaning up variable states.
2014-01-30 21:26:33.403 openEarsTest[1700:4107] Processing speech, please wait…
INFO: file_omitted(0): cmn_prior_update: from < 47.00 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: file_omitted(0): cmn_prior_update: to < 52.53 -1.62 -2.63 -4.43 -4.27 -1.81 -0.63 0.20 0.11 -0.32 -0.38 -0.34 -0.05 >
INFO: file_omitted(0): 989 words recognized (9/fr)
INFO: file_omitted(0): 5292 senones evaluated (49/fr)
INFO: file_omitted(0): 2026 channels searched (18/fr), 103 1st, 1645 last
INFO: file_omitted(0): 1107 words for which last channels evaluated (10/fr)
INFO: file_omitted(0): 24 candidate words for entering last phone (0/fr)
INFO: file_omitted(0): fwdtree 0.08 CPU 0.072 xRT
INFO: file_omitted(0): fwdtree 1.87 wall 1.749 xRT
INFO: file_omitted(0): Utterance vocabulary contains 2 words
INFO: file_omitted(0): 185 words recognized (2/fr)
INFO: file_omitted(0): 6188 senones evaluated (58/fr)
INFO: file_omitted(0): 3642 channels searched (34/fr)
INFO: file_omitted(0): 269 words searched (2/fr)
INFO: file_omitted(0): 76 word transitions (0/fr)
INFO: file_omitted(0): fwdflat 0.03 CPU 0.028 xRT
INFO: file_omitted(0): fwdflat 0.03 wall 0.027 xRT
INFO: file_omitted(0): </s> not found in last frame, using OKDEAR.105 instead
INFO: file_omitted(0): lattice start node <s>.0 end node OKDEAR.25
INFO: file_omitted(0): Eliminated 0 nodes before end node
INFO: file_omitted(0): Lattice has 19 nodes, 2 links
INFO: file_omitted(0): Normalizer P(O) = alpha(OKDEAR:25:105) = -852341
INFO: file_omitted(0): Joint P(O,S) = -852341 P(S|O) = 0
INFO: file_omitted(0): bestpath 0.00 CPU 0.000 xRT
INFO: file_omitted(0): bestpath 0.00 wall 0.000 xRT
2014-01-30 21:26:33.436 openEarsTest[1700:4107] Pocketsphinx heard “OKDEAR” with a score of (0) and an utterance ID of 000000000.
2014-01-30 21:26:33.438 openEarsTest[1700:60b] The received hypothesis is OKDEAR with a score of 0 and an ID of 000000000
2014-01-30 21:26:33.438 openEarsTest[1700:4107] Checking and resetting all audio session settings.
2014-01-30 21:26:33.441 openEarsTest[1700:4107] audioCategory is correct, we will leave it as it is.
2014-01-30 21:26:33.442 openEarsTest[1700:4107] bluetoothInput is correct, we will leave it as it is.
2014-01-30 21:26:33.443 openEarsTest[1700:4107] Output Device: SpeakerAndMicrophone.
2014-01-30 21:26:33.444 openEarsTest[1700:4107] categoryDefaultToSpeaker is correct, we will leave it as it is.
2014-01-30 21:26:33.446 openEarsTest[1700:4107] preferredBufferSize is correct, we will leave it as it is.
2014-01-30 21:26:33.447 openEarsTest[1700:4107] preferredSampleRateCheck is correct, we will leave it as it is.
2014-01-30 21:26:33.448 openEarsTest[1700:4107] Setting the variables for the device and starting it.
2014-01-30 21:26:33.450 openEarsTest[1700:4107] Looping through ringbuffer sections and pre-allocating them.
2014-01-30 21:26:33.526 openEarsTest[1700:4107] Started audio output unit.
2014-01-30 21:26:33.528 openEarsTest[1700:4107] Listening.
2014-01-30 21:26:33.529 openEarsTest[1700:60b] Pocketsphinx is now listening.
2014-01-30 21:26:45.993 openEarsTest[1700:4107] Speech detected…
2014-01-30 21:26:45.996 openEarsTest[1700:60b] Pocketsphinx has detected speech.
INFO: file_omitted(0): Resized backpointer table to 10000 entries
2014-01-30 21:26:50.944 openEarsTest[1700:4107] There is reason to suspect the VAD of being out of sync with the current background noise levels in the environment so we will recalibrate.
2014-01-30 21:26:50.946 openEarsTest[1700:60b] Pocketsphinx has detected a period of silence, concluding an utterance.
2014-01-30 21:26:50.946 openEarsTest[1700:4107] Stopping audio unit.
2014-01-30 21:26:51.066 openEarsTest[1700:4107] Audio Output Unit stopped, cleaning up variable states.
2014-01-30 21:26:51.067 openEarsTest[1700:4107] Processing speech, please wait…
INFO: file_omitted(0): cmn_prior_update: from < 52.53 -1.62 -2.63 -4.43 -4.27 -1.81 -0.63 0.20 0.11 -0.32 -0.38 -0.34 -0.05 >
INFO: file_omitted(0): cmn_prior_update: to < 54.96 -1.75 -3.88 -4.73 -2.06 -3.35 -1.31 0.39 -0.73 0.04 0.32 -0.15 0.14 >
INFO: file_omitted(0): 5144 words recognized (10/fr)
INFO: file_omitted(0): 23068 senones evaluated (45/fr)
INFO: file_omitted(0): 7218 channels searched (13/fr), 514 1st, 5197 last
INFO: file_omitted(0): 5197 words for which last channels evaluated (10/fr)
INFO: file_omitted(0): 40 candidate words for entering last phone (0/fr)
INFO: file_omitted(0): fwdtree 0.33 CPU 0.064 xRT
INFO: file_omitted(0): fwdtree 5.07 wall 0.980 xRT
INFO: file_omitted(0): Utterance vocabulary contains 1 words
INFO: file_omitted(0): 540 words recognized (1/fr)
INFO: file_omitted(0): 1551 senones evaluated (3/fr)
INFO: file_omitted(0): 588 channels searched (1/fr)
INFO: file_omitted(0): 588 words searched (1/fr)
INFO: file_omitted(0): 26 word transitions (0/fr)
INFO: file_omitted(0): fwdflat 0.10 CPU 0.020 xRT
INFO: file_omitted(0): fwdflat 0.11 wall 0.020 xRT
INFO: file_omitted(0): </s> not found in last frame, using <sil>.516 instead
INFO: file_omitted(0): lattice start node <s>.0 end node <sil>.493
INFO: file_omitted(0): Eliminated 0 nodes before end node
INFO: file_omitted(0): Lattice has 27 nodes, 5 links
INFO: file_omitted(0): Normalizer P(O) = alpha(<sil>:493:516) = -3680852
INFO: file_omitted(0): Joint P(O,S) = -3680852 P(S|O) = 0
INFO: file_omitted(0): bestpath 0.00 CPU 0.000 xRT
INFO: file_omitted(0): bestpath 0.00 wall 0.000 xRT
2014-01-30 21:26:51.177 openEarsTest[1700:4107] Pocketsphinx heard “” with a score of (0) and an utterance ID of 000000001.
2014-01-30 21:26:51.179 openEarsTest[1700:4107] Hypothesis was null so we aren’t returning it. If you want null hypotheses to also be returned, set PocketsphinxController’s property returnNullHypotheses to TRUE before starting PocketsphinxController.
2014-01-30 21:26:51.181 openEarsTest[1700:4107] Checking and resetting all audio session settings.
2014-01-30 21:26:51.182 openEarsTest[1700:4107] audioCategory is correct, we will leave it as it is.
2014-01-30 21:26:51.184 openEarsTest[1700:4107] bluetoothInput is correct, we will leave it as it is.
2014-01-30 21:26:51.185 openEarsTest[1700:4107] Output Device: SpeakerAndMicrophone.
2014-01-30 21:26:51.187 openEarsTest[1700:4107] categoryDefaultToSpeaker is correct, we will leave it as it is.
2014-01-30 21:26:51.188 openEarsTest[1700:4107] preferredBufferSize is correct, we will leave it as it is.
2014-01-30 21:26:51.189 openEarsTest[1700:4107] preferredSampleRateCheck is correct, we will leave it as it is.
2014-01-30 21:26:51.191 openEarsTest[1700:4107] Setting the variables for the device and starting it.
2014-01-30 21:26:51.192 openEarsTest[1700:4107] Looping through ringbuffer sections and pre-allocating them.
2014-01-30 21:26:51.319 openEarsTest[1700:4107] Started audio output unit.
2014-01-30 21:26:51.321 openEarsTest[1700:4107] Listening.
2014-01-30 21:26:51.323 openEarsTest[1700:60b] Pocketsphinx is now listening.
2014-01-30 21:26:51.851 openEarsTest[1700:4107] Speech detected…
2014-01-30 21:26:51.853 openEarsTest[1700:60b] Pocketsphinx has detected speech.
INFO: file_omitted(0): Resized backpointer table to 20000 entries
2014-01-30 21:27:01.696 openEarsTest[1700:4107] There is reason to suspect the VAD of being out of sync with the current background noise levels in the environment so we will recalibrate.
2014-01-30 21:27:01.697 openEarsTest[1700:60b] Pocketsphinx has detected a period of silence, concluding an utterance.
2014-01-30 21:27:01.697 openEarsTest[1700:4107] Stopping audio unit.
2014-01-30 21:27:01.817 openEarsTest[1700:4107] Audio Output Unit stopped, cleaning up variable states.
2014-01-30 21:27:01.819 openEarsTest[1700:4107] Processing speech, please wait…
INFO: file_omitted(0): cmn_prior_update: from < 55.97 -1.72 -4.37 -5.17 -1.74 -3.51 -1.12 0.46 -0.76 0.14 0.54 -0.12 0.05 >
INFO: file_omitted(0): cmn_prior_update: to < 55.70 -1.84 -4.25 -5.14 -1.75 -3.60 -1.18 0.29 -0.66 0.16 0.39 -0.07 0.09 >
INFO: file_omitted(0): 10064 words recognized (10/fr)
INFO: file_omitted(0): 44995 senones evaluated (45/fr)
INFO: file_omitted(0): 14056 channels searched (13/fr), 1001 1st, 10165 last
INFO: file_omitted(0): 10099 words for which last channels evaluated (10/fr)
INFO: file_omitted(0): 91 candidate words for entering last phone (0/fr)
INFO: file_omitted(0): fwdtree 0.64 CPU 0.064 xRT
INFO: file_omitted(0): fwdtree 9.97 wall 0.992 xRT
INFO: file_omitted(0): Utterance vocabulary contains 1 words
INFO: file_omitted(0): 1027 words recognized (1/fr)
INFO: file_omitted(0): 3012 senones evaluated (3/fr)
INFO: file_omitted(0): 1077 channels searched (1/fr)
INFO: file_omitted(0): 1077 words searched (1/fr)
INFO: file_omitted(0): 26 word transitions (0/fr)
INFO: file_omitted(0): fwdflat 0.19 CPU 0.019 xRT
INFO: file_omitted(0): fwdflat 0.19 wall 0.019 xRT
INFO: file_omitted(0): </s> not found in last frame, using <sil>.1003 instead
INFO: file_omitted(0): lattice start node <s>.0 end node <sil>.818
INFO: file_omitted(0): Eliminated 0 nodes before end node
INFO: file_omitted(0): Lattice has 25 nodes, 5 links
INFO: file_omitted(0): Normalizer P(O) = alpha(<sil>:818:1003) = -6299169
INFO: file_omitted(0): Joint P(O,S) = -6299169 P(S|O) = 0
INFO: file_omitted(0): bestpath 0.00 CPU 0.000 xRT
INFO: file_omitted(0): bestpath 0.00 wall 0.000 xRT
2014-01-30 21:27:02.012 openEarsTest[1700:4107] Pocketsphinx heard “” with a score of (0) and an utterance ID of 000000002.
2014-01-30 21:27:02.013 openEarsTest[1700:4107] Hypothesis was null so we aren’t returning it. If you want null hypotheses to also be returned, set PocketsphinxController’s property returnNullHypotheses to TRUE before starting PocketsphinxController.
2014-01-30 21:27:02.015 openEarsTest[1700:4107] Checking and resetting all audio session settings.
2014-01-30 21:27:02.016 openEarsTest[1700:4107] audioCategory is correct, we will leave it as it is.
2014-01-30 21:27:02.017 openEarsTest[1700:4107] bluetoothInput is correct, we will leave it as it is.
2014-01-30 21:27:02.018 openEarsTest[1700:4107] Output Device: SpeakerAndMicrophone.
2014-01-30 21:27:02.019 openEarsTest[1700:4107] categoryDefaultToSpeaker is correct, we will leave it as it is.
2014-01-30 21:27:02.020 openEarsTest[1700:4107] preferredBufferSize is correct, we will leave it as it is.
2014-01-30 21:27:02.021 openEarsTest[1700:4107] preferredSampleRateCheck is correct, we will leave it as it is.
2014-01-30 21:27:02.022 openEarsTest[1700:4107] Setting the variables for the device and starting it.
2014-01-30 21:27:02.023 openEarsTest[1700:4107] Looping through ringbuffer sections and pre-allocating them.
2014-01-30 21:27:02.071 openEarsTest[1700:4107] Started audio output unit.
2014-01-30 21:27:02.073 openEarsTest[1700:4107] Listening.
2014-01-30 21:27:02.075 openEarsTest[1700:60b] Pocketsphinx is now listening.
2014-01-30 21:27:02.602 openEarsTest[1700:4107] Speech detected…
2014-01-30 21:27:02.605 openEarsTest[1700:60b] Pocketsphinx has detected speech.
2014-01-30 21:27:09.754 openEarsTest[1700:4107] Stopping audio unit.
2014-01-30 21:27:09.755 openEarsTest[1700:60b] Pocketsphinx has detected a period of silence, concluding an utterance.
2014-01-30 21:27:09.882 openEarsTest[1700:4107] Audio Output Unit stopped, cleaning up variable states.
2014-01-30 21:27:09.883 openEarsTest[1700:4107] Processing speech, please wait…
INFO: file_omitted(0): cmn_prior_update: from < 55.98 -1.77 -4.29 -5.13 -1.61 -3.50 -1.24 0.45 -0.59 0.11 0.32 -0.28 0.06 >
INFO: file_omitted(0): cmn_prior_update: to < 55.99 -1.98 -4.25 -4.92 -1.83 -3.69 -1.24 0.42 -0.76 0.21 0.51 0.06 0.17 >
INFO: file_omitted(0): 6560 words recognized (10/fr)
INFO: file_omitted(0): 29301 senones evaluated (45/fr)
INFO: file_omitted(0): 9187 channels searched (14/fr), 652 1st, 6636 last
INFO: file_omitted(0): 6603 words for which last channels evaluated (10/fr)
INFO: file_omitted(0): 71 candidate words for entering last phone (0/fr)
INFO: file_omitted(0): fwdtree 0.42 CPU 0.065 xRT
INFO: file_omitted(0): fwdtree 7.28 wall 1.110 xRT
INFO: file_omitted(0): Utterance vocabulary contains 1 words
INFO: file_omitted(0): 677 words recognized (1/fr)
INFO: file_omitted(0): 1965 senones evaluated (3/fr)
INFO: file_omitted(0): 724 channels searched (1/fr)
INFO: file_omitted(0): 724 words searched (1/fr)
INFO: file_omitted(0): 26 word transitions (0/fr)
INFO: file_omitted(0): fwdflat 0.12 CPU 0.019 xRT
INFO: file_omitted(0): fwdflat 0.13 wall 0.020 xRT
INFO: file_omitted(0): </s> not found in last frame, using <sil>.654 instead
INFO: file_omitted(0): lattice start node <s>.0 end node <sil>.618
INFO: file_omitted(0): Eliminated 0 nodes before end node
INFO: file_omitted(0): Lattice has 25 nodes, 4 links
INFO: file_omitted(0): Normalizer P(O) = alpha(<sil>:618:654) = -4297454
INFO: file_omitted(0): Joint P(O,S) = -4297455 P(S|O) = -1
INFO: file_omitted(0): bestpath 0.00 CPU 0.000 xRT
INFO: file_omitted(0): bestpath 0.00 wall 0.000 xRT
2014-01-30 21:27:10.015 openEarsTest[1700:4107] Pocketsphinx heard “” with a score of (-1) and an utterance ID of 000000003.
2014-01-30 21:27:10.017 openEarsTest[1700:4107] Hypothesis was null so we aren’t returning it. If you want null hypotheses to also be returned, set PocketsphinxController’s property returnNullHypotheses to TRUE before starting PocketsphinxController.
2014-01-30 21:27:10.018 openEarsTest[1700:4107] Checking and resetting all audio session settings.
2014-01-30 21:27:10.020 openEarsTest[1700:4107] audioCategory is correct, we will leave it as it is.
2014-01-30 21:27:10.021 openEarsTest[1700:4107] bluetoothInput is correct, we will leave it as it is.
2014-01-30 21:27:10.022 openEarsTest[1700:4107] Output Device: SpeakerAndMicrophone.
2014-01-30 21:27:10.023 openEarsTest[1700:4107] categoryDefaultToSpeaker is correct, we will leave it as it is.
2014-01-30 21:27:10.024 openEarsTest[1700:4107] preferredBufferSize is correct, we will leave it as it is.
2014-01-30 21:27:10.025 openEarsTest[1700:4107] preferredSampleRateCheck is correct, we will leave it as it is.
2014-01-30 21:27:10.026 openEarsTest[1700:4107] Setting the variables for the device and starting it.
2014-01-30 21:27:10.027 openEarsTest[1700:4107] Looping through ringbuffer sections and pre-allocating them.
2014-01-30 21:27:10.135 openEarsTest[1700:4107] Started audio output unit.
2014-01-30 21:27:10.137 openEarsTest[1700:4107] Listening.
2014-01-30 21:27:10.139 openEarsTest[1700:60b] Pocketsphinx is now listening.
2014-01-30 21:27:20.791 openEarsTest[1700:4107] Speech detected…
2014-01-30 21:27:20.794 openEarsTest[1700:60b] Pocketsphinx has detected speech.
2014-01-30 21:27:23.448 openEarsTest[1700:4107] Stopping audio unit.
2014-01-30 21:27:23.448 openEarsTest[1700:60b] Pocketsphinx has detected a period of silence, concluding an utterance.
2014-01-30 21:27:23.577 openEarsTest[1700:4107] Audio Output Unit stopped, cleaning up variable states.
2014-01-30 21:27:23.579 openEarsTest[1700:4107] Processing speech, please wait…
INFO: file_omitted(0): cmn_prior_update: from < 56.08 -1.88 -4.29 -5.02 -1.83 -3.60 -1.21 0.47 -0.74 0.23 0.48 0.04 0.20 >
INFO: file_omitted(0): cmn_prior_update: to < 57.60 -1.74 -4.48 -4.72 -1.80 -3.61 -1.10 0.53 -0.64 0.13 0.15 -0.28 0.16 >
INFO: file_omitted(0): 1993 words recognized (10/fr)
INFO: file_omitted(0): 9583 senones evaluated (47/fr)
INFO: file_omitted(0): 3323 channels searched (16/fr), 199 1st, 2554 last
INFO: file_omitted(0): 2058 words for which last channels evaluated (10/fr)
INFO: file_omitted(0): 70 candidate words for entering last phone (0/fr)
INFO: file_omitted(0): fwdtree 0.14 CPU 0.069 xRT
INFO: file_omitted(0): fwdtree 2.79 wall 1.374 xRT
INFO: file_omitted(0): Utterance vocabulary contains 2 words
INFO: file_omitted(0): 350 words recognized (2/fr)
INFO: file_omitted(0): 10478 senones evaluated (52/fr)
INFO: file_omitted(0): 5804 channels searched (28/fr)
INFO: file_omitted(0): 458 words searched (2/fr)
INFO: file_omitted(0): 76 word transitions (0/fr)
INFO: file_omitted(0): fwdflat 0.05 CPU 0.025 xRT
INFO: file_omitted(0): fwdflat 0.05 wall 0.027 xRT
INFO: file_omitted(0): </s> not found in last frame, using <sil>.201 instead
INFO: file_omitted(0): lattice start node <s>.0 end node <sil>.103
INFO: file_omitted(0): Eliminated 0 nodes before end node
INFO: file_omitted(0): Lattice has 20 nodes, 6 links
INFO: file_omitted(0): Normalizer P(O) = alpha(<sil>:103:201) = -1576540
INFO: file_omitted(0): Joint P(O,S) = -1577862 P(S|O) = -1322
INFO: file_omitted(0): bestpath 0.00 CPU 0.000 xRT
INFO: file_omitted(0): bestpath 0.00 wall 0.000 xRT
2014-01-30 21:27:23.637 openEarsTest[1700:4107] Pocketsphinx heard “OKDEAR” with a score of (-1322) and an utterance ID of 000000004.
2014-01-30 21:27:23.639 openEarsTest[1700:4107] Checking and resetting all audio session settings.
2014-01-30 21:27:23.639 openEarsTest[1700:60b] The received hypothesis is OKDEAR with a score of -1322 and an ID of 000000004
2014-01-30 21:27:23.641 openEarsTest[1700:4107] audioCategory is correct, we will leave it as it is.
2014-01-30 21:27:23.642 openEarsTest[1700:4107] bluetoothInput is correct, we will leave it as it is.
2014-01-30 21:27:23.644 openEarsTest[1700:4107] Output Device: SpeakerAndMicrophone.
2014-01-30 21:27:23.645 openEarsTest[1700:4107] categoryDefaultToSpeaker is correct, we will leave it as it is.
2014-01-30 21:27:23.646 openEarsTest[1700:4107] preferredBufferSize is correct, we will leave it as it is.
2014-01-30 21:27:23.647 openEarsTest[1700:4107] preferredSampleRateCheck is correct, we will leave it as it is.
2014-01-30 21:27:23.649 openEarsTest[1700:4107] Setting the variables for the device and starting it.
2014-01-30 21:27:23.650 openEarsTest[1700:4107] Looping through ringbuffer sections and pre-allocating them.
2014-01-30 21:27:23.703 openEarsTest[1700:4107] Started audio output unit.
2014-01-30 21:27:23.705 openEarsTest[1700:4107] Listening.
2014-01-30 21:27:23.707 openEarsTest[1700:60b] Pocketsphinx is now listening.
2014-01-30 21:27:35.783 openEarsTest[1700:4107] Speech detected…
2014-01-30 21:27:35.785 openEarsTest[1700:60b] Pocketsphinx has detected speech.
2014-01-30 21:27:37.528 openEarsTest[1700:4107] Stopping audio unit.
2014-01-30 21:27:37.528 openEarsTest[1700:60b] Pocketsphinx has detected a period of silence, concluding an utterance.
2014-01-30 21:27:37.656 openEarsTest[1700:4107] Audio Output Unit stopped, cleaning up variable states.
2014-01-30 21:27:37.659 openEarsTest[1700:4107] Processing speech, please wait…
INFO: file_omitted(0): cmn_prior_update: from < 57.60 -1.74 -4.48 -4.72 -1.80 -3.61 -1.10 0.53 -0.64 0.13 0.15 -0.28 0.16 >
INFO: file_omitted(0): cmn_prior_update: to < 58.42 -1.84 -4.41 -4.80 -2.09 -3.41 -1.04 0.52 -0.45 0.09 0.09 -0.19 0.18 >
INFO: file_omitted(0): 954 words recognized (9/fr)
INFO: file_omitted(0): 4555 senones evaluated (45/fr)
INFO: file_omitted(0): 1561 channels searched (15/fr), 97 1st, 1195 last
INFO: file_omitted(0): 1027 words for which last channels evaluated (10/fr)
INFO: file_omitted(0): 37 candidate words for entering last phone (0/fr)
INFO: file_omitted(0): fwdtree 0.07 CPU 0.071 xRT
INFO: file_omitted(0): fwdtree 1.88 wall 1.858 xRT
INFO: file_omitted(0): Utterance vocabulary contains 1 words
INFO: file_omitted(0): 115 words recognized (1/fr)
INFO: file_omitted(0): 300 senones evaluated (3/fr)
INFO: file_omitted(0): 163 channels searched (1/fr)
INFO: file_omitted(0): 163 words searched (1/fr)
INFO: file_omitted(0): 26 word transitions (0/fr)
INFO: file_omitted(0): fwdflat 0.02 CPU 0.020 xRT
INFO: file_omitted(0): fwdflat 0.02 wall 0.021 xRT
INFO: file_omitted(0): </s> not found in last frame, using <sil>.99 instead
INFO: file_omitted(0): lattice start node <s>.0 end node <sil>.80
INFO: file_omitted(0): Eliminated 0 nodes before end node
INFO: file_omitted(0): Lattice has 15 nodes, 2 links
INFO: file_omitted(0): Normalizer P(O) = alpha(<sil>:80:99) = -806075
INFO: file_omitted(0): Joint P(O,S) = -806075 P(S|O) = 0
INFO: file_omitted(0): bestpath 0.00 CPU 0.000 xRT
INFO: file_omitted(0): bestpath 0.00 wall 0.000 xRT
2014-01-30 21:27:37.683 openEarsTest[1700:4107] Pocketsphinx heard “” with a score of (0) and an utterance ID of 000000005.
2014-01-30 21:27:37.684 openEarsTest[1700:4107] Hypothesis was null so we aren’t returning it. If you want null hypotheses to also be returned, set PocketsphinxController’s property returnNullHypotheses to TRUE before starting PocketsphinxController.
2014-01-30 21:27:37.686 openEarsTest[1700:4107] Checking and resetting all audio session settings.
2014-01-30 21:27:37.687 openEarsTest[1700:4107] audioCategory is correct, we will leave it as it is.
2014-01-30 21:27:37.689 openEarsTest[1700:4107] bluetoothInput is correct, we will leave it as it is.
2014-01-30 21:27:37.690 openEarsTest[1700:4107] Output Device: SpeakerAndMicrophone.
2014-01-30 21:27:37.691 openEarsTest[1700:4107] categoryDefaultToSpeaker is correct, we will leave it as it is.
2014-01-30 21:27:37.692 openEarsTest[1700:4107] preferredBufferSize is correct, we will leave it as it is.
2014-01-30 21:27:37.694 openEarsTest[1700:4107] preferredSampleRateCheck is correct, we will leave it as it is.
2014-01-30 21:27:37.695 openEarsTest[1700:4107] Setting the variables for the device and starting it.
2014-01-30 21:27:37.696 openEarsTest[1700:4107] Looping through ringbuffer sections and pre-allocating them.
2014-01-30 21:27:37.783 openEarsTest[1700:4107] Started audio output unit.
2014-01-30 21:27:37.785 openEarsTest[1700:4107] Listening.
2014-01-30 21:27:37.786 openEarsTest[1700:60b] Pocketsphinx is now listening.
2014-01-30 21:27:45.366 openEarsTest[1700:4107] Speech detected…
2014-01-30 21:27:45.368 openEarsTest[1700:60b] Pocketsphinx has detected speech.
2014-01-30 21:27:47.386 openEarsTest[1700:4107] Stopping audio unit.
2014-01-30 21:27:47.386 openEarsTest[1700:60b] Pocketsphinx has detected a period of silence, concluding an utterance.
2014-01-30 21:27:47.514 openEarsTest[1700:4107] Audio Output Unit stopped, cleaning up variable states.
2014-01-30 21:27:47.515 openEarsTest[1700:4107] Processing speech, please wait…
INFO: file_omitted(0): cmn_prior_update: from < 58.49 -1.82 -4.37 -4.91 -2.10 -3.32 -1.06 0.52 -0.45 0.10 0.11 -0.17 0.14 >
INFO: file_omitted(0): cmn_prior_update: to < 59.80 -1.90 -4.38 -4.94 -2.47 -3.00 -0.89 0.54 -0.21 0.03 0.03 -0.17 0.14 >
INFO: file_omitted(0): 1275 words recognized (10/fr)
INFO: file_omitted(0): 6133 senones evaluated (46/fr)
INFO: file_omitted(0): 2120 channels searched (15/fr), 129 1st, 1635 last
INFO: file_omitted(0): 1363 words for which last channels evaluated (10/fr)
INFO: file_omitted(0): 27 candidate words for entering last phone (0/fr)
INFO: file_omitted(0): fwdtree 0.09 CPU 0.069 xRT
INFO: file_omitted(0): fwdtree 2.15 wall 1.616 xRT
INFO: file_omitted(0): Utterance vocabulary contains 1 words
INFO: file_omitted(0): 151 words recognized (1/fr)
INFO: file_omitted(0): 396 senones evaluated (3/fr)
INFO: file_omitted(0): 195 channels searched (1/fr)
INFO: file_omitted(0): 195 words searched (1/fr)
INFO: file_omitted(0): 26 word transitions (0/fr)
INFO: file_omitted(0): fwdflat 0.02 CPU 0.015 xRT
INFO: file_omitted(0): fwdflat 0.03 wall 0.021 xRT
INFO: file_omitted(0): </s> not found in last frame, using <sil>.131 instead
INFO: file_omitted(0): lattice start node <s>.0 end node <sil>.103
INFO: file_omitted(0): Eliminated 0 nodes before end node
INFO: file_omitted(0): Lattice has 20 nodes, 2 links
INFO: file_omitted(0): Normalizer P(O) = alpha(<sil>:103:131) = -1038369
INFO: file_omitted(0): Joint P(O,S) = -1038369 P(S|O) = 0
INFO: file_omitted(0): bestpath 0.00 CPU 0.000 xRT
INFO: file_omitted(0): bestpath 0.00 wall 0.000 xRT
2014-01-30 21:27:47.545 openEarsTest[1700:4107] Pocketsphinx heard “” with a score of (0) and an utterance ID of 000000006.
2014-01-30 21:27:47.547 openEarsTest[1700:4107] Hypothesis was null so we aren’t returning it. If you want null hypotheses to also be returned, set PocketsphinxController’s property returnNullHypotheses to TRUE before starting PocketsphinxController.
2014-01-30 21:27:47.548 openEarsTest[1700:4107] Checking and resetting all audio session settings.
2014-01-30 21:27:47.550 openEarsTest[1700:4107] audioCategory is correct, we will leave it as it is.
2014-01-30 21:27:47.552 openEarsTest[1700:4107] bluetoothInput is correct, we will leave it as it is.
2014-01-30 21:27:47.553 openEarsTest[1700:4107] Output Device: SpeakerAndMicrophone.
2014-01-30 21:27:47.554 openEarsTest[1700:4107] categoryDefaultToSpeaker is correct, we will leave it as it is.
2014-01-30 21:27:47.556 openEarsTest[1700:4107] preferredBufferSize is correct, we will leave it as it is.
2014-01-30 21:27:47.557 openEarsTest[1700:4107] preferredSampleRateCheck is correct, we will leave it as it is.
2014-01-30 21:27:47.558 openEarsTest[1700:4107] Setting the variables for the device and starting it.
2014-01-30 21:27:47.559 openEarsTest[1700:4107] Looping through ringbuffer sections and pre-allocating them.
2014-01-30 21:27:47.639 openEarsTest[1700:4107] Started audio output unit.
2014-01-30 21:27:47.641 openEarsTest[1700:4107] Listening.
2014-01-30 21:27:47.643 openEarsTest[1700:60b] Pocketsphinx is now listening.
2014-01-30 21:27:53.575 openEarsTest[1700:4107] Speech detected…
2014-01-30 21:27:53.577 openEarsTest[1700:60b] Pocketsphinx has detected speech.
2014-01-30 21:27:55.323 openEarsTest[1700:4107] Stopping audio unit.
2014-01-30 21:27:55.323 openEarsTest[1700:60b] Pocketsphinx has detected a period of silence, concluding an utterance.
2014-01-30 21:27:55.450 openEarsTest[1700:4107] Audio Output Unit stopped, cleaning up variable states.
2014-01-30 21:27:55.452 openEarsTest[1700:4107] Processing speech, please wait…
INFO: file_omitted(0): cmn_prior_update: from < 59.80 -1.90 -4.38 -4.94 -2.47 -3.00 -0.89 0.54 -0.21 0.03 0.03 -0.17 0.14 >
INFO: file_omitted(0): cmn_prior_update: to < 60.54 -2.01 -4.31 -5.02 -2.58 -2.79 -0.78 0.50 -0.15 -0.05 -0.00 -0.17 0.11 >
INFO: file_omitted(0): 1088 words recognized (10/fr)
INFO: file_omitted(0): 5443 senones evaluated (48/fr)
INFO: file_omitted(0): 1981 channels searched (17/fr), 110 1st, 1561 last
INFO: file_omitted(0): 1165 words for which last channels evaluated (10/fr)
INFO: file_omitted(0): 41 candidate words for entering last phone (0/fr)
INFO: file_omitted(0): fwdtree 0.08 CPU 0.070 xRT
INFO: file_omitted(0): fwdtree 1.88 wall 1.647 xRT
INFO: file_omitted(0): Utterance vocabulary contains 1 words
INFO: file_omitted(0): 128 words recognized (1/fr)
INFO: file_omitted(0): 339 senones evaluated (3/fr)
INFO: file_omitted(0): 180 channels searched (1/fr)
INFO: file_omitted(0): 180 words searched (1/fr)
INFO: file_omitted(0): 26 word transitions (0/fr)
INFO: file_omitted(0): fwdflat 0.02 CPU 0.017 xRT
INFO: file_omitted(0): fwdflat 0.02 wall 0.020 xRT
INFO: file_omitted(0): </s> not found in last frame, using <sil>.112 instead
INFO: file_omitted(0): lattice start node <s>.0 end node <sil>.82
INFO: file_omitted(0): Eliminated 0 nodes before end node
INFO: file_omitted(0): Lattice has 15 nodes, 2 links
INFO: file_omitted(0): Normalizer P(O) = alpha(<sil>:82:112) = -888199
INFO: file_omitted(0): Joint P(O,S) = -888199 P(S|O) = 0
INFO: file_omitted(0): bestpath 0.00 CPU 0.000 xRT
INFO: file_omitted(0): bestpath 0.00 wall 0.000 xRT
2014-01-30 21:27:55.478 openEarsTest[1700:4107] Pocketsphinx heard “” with a score of (0) and an utterance ID of 000000007.
2014-01-30 21:27:55.479 openEarsTest[1700:4107] Hypothesis was null so we aren’t returning it. If you want null hypotheses to also be returned, set PocketsphinxController’s property returnNullHypotheses to TRUE before starting PocketsphinxController.
2014-01-30 21:27:55.481 openEarsTest[1700:4107] Checking and resetting all audio session settings.
2014-01-30 21:27:55.483 openEarsTest[1700:4107] audioCategory is correct, we will leave it as it is.
2014-01-30 21:27:55.484 openEarsTest[1700:4107] bluetoothInput is correct, we will leave it as it is.
2014-01-30 21:27:55.485 openEarsTest[1700:4107] Output Device: SpeakerAndMicrophone.
2014-01-30 21:27:55.486 openEarsTest[1700:4107] categoryDefaultToSpeaker is correct, we will leave it as it is.
2014-01-30 21:27:55.487 openEarsTest[1700:4107] preferredBufferSize is correct, we will leave it as it is.
2014-01-30 21:27:55.489 openEarsTest[1700:4107] preferredSampleRateCheck is correct, we will leave it as it is.
2014-01-30 21:27:55.490 openEarsTest[1700:4107] Setting the variables for the device and starting it.
2014-01-30 21:27:55.491 openEarsTest[1700:4107] Looping through ringbuffer sections and pre-allocating them.
2014-01-30 21:27:55.575 openEarsTest[1700:4107] Started audio output unit.
2014-01-30 21:27:55.577 openEarsTest[1700:4107] Listening.
2014-01-30 21:27:55.579 openEarsTest[1700:60b] Pocketsphinx is now listening.
2014-01-30 21:28:03.561 openEarsTest[1700:4107] Speech detected…
2014-01-30 21:28:03.568 openEarsTest[1700:60b] Pocketsphinx has detected speech.
2014-01-30 21:28:05.564 openEarsTest[1700:4107] Stopping audio unit.
2014-01-30 21:28:05.564 openEarsTest[1700:60b] Pocketsphinx has detected a period of silence, concluding an utterance.
2014-01-30 21:28:05.690 openEarsTest[1700:4107] Audio Output Unit stopped, cleaning up variable states.
2014-01-30 21:28:05.691 openEarsTest[1700:4107] Processing speech, please wait…
INFO: file_omitted(0): cmn_prior_update: from < 61.00 -2.09 -4.21 -5.11 -2.72 -2.55 -0.79 0.50 -0.05 -0.11 -0.03 -0.21 0.09 >
INFO: file_omitted(0): cmn_prior_update: to < 61.45 -2.18 -4.31 -5.13 -2.74 -2.52 -0.69 0.55 -0.01 -0.16 -0.04 -0.17 0.06 >
INFO: file_omitted(0): 1306 words recognized (10/fr)
INFO: file_omitted(0): 6301 senones evaluated (46/fr)
INFO: file_omitted(0): 2198 channels searched (16/fr), 132 1st, 1703 last
INFO: file_omitted(0): 1400 words for which last channels evaluated (10/fr)
INFO: file_omitted(0): 26 candidate words for entering last phone (0/fr)
INFO: file_omitted(0): fwdtree 0.09 CPU 0.069 xRT
INFO: file_omitted(0): fwdtree 2.13 wall 1.567 xRT
INFO: file_omitted(0): Utterance vocabulary contains 1 words
INFO: file_omitted(0): 156 words recognized (1/fr)
INFO: file_omitted(0): 405 senones evaluated (3/fr)
INFO: file_omitted(0): 198 channels searched (1/fr)
INFO: file_omitted(0): 198 words searched (1/fr)
INFO: file_omitted(0): 26 word transitions (0/fr)
INFO: file_omitted(0): fwdflat 0.02 CPU 0.015 xRT
INFO: file_omitted(0): fwdflat 0.03 wall 0.020 xRT
INFO: file_omitted(0): </s> not found in last frame, using <sil>.134 instead
INFO: file_omitted(0): lattice start node <s>.0 end node <sil>.113
INFO: file_omitted(0): Eliminated 0 nodes before end node
INFO: file_omitted(0): Lattice has 20 nodes, 2 links
INFO: file_omitted(0): Normalizer P(O) = alpha(<sil>:113:134) = -1010670
INFO: file_omitted(0): Joint P(O,S) = -1010670 P(S|O) = 0
INFO: file_omitted(0): bestpath 0.00 CPU 0.000 xRT
INFO: file_omitted(0): bestpath 0.00 wall 0.000 xRT
2014-01-30 21:28:05.720 openEarsTest[1700:4107] Pocketsphinx heard “” with a score of (0) and an utterance ID of 000000008.
2014-01-30 21:28:05.722 openEarsTest[1700:4107] Hypothesis was null so we aren’t returning it. If you want null hypotheses to also be returned, set PocketsphinxController’s property returnNullHypotheses to TRUE before starting PocketsphinxController.
2014-01-30 21:28:05.723 openEarsTest[1700:4107] Checking and resetting all audio session settings.
2014-01-30 21:28:05.725 openEarsTest[1700:4107] audioCategory is correct, we will leave it as it is.
2014-01-30 21:28:05.726 openEarsTest[1700:4107] bluetoothInput is correct, we will leave it as it is.
2014-01-30 21:28:05.727 openEarsTest[1700:4107] Output Device: SpeakerAndMicrophone.
2014-01-30 21:28:05.729 openEarsTest[1700:4107] categoryDefaultToSpeaker is correct, we will leave it as it is.
2014-01-30 21:28:05.730 openEarsTest[1700:4107] preferredBufferSize is correct, we will leave it as it is.
2014-01-30 21:28:05.731 openEarsTest[1700:4107] preferredSampleRateCheck is correct, we will leave it as it is.
2014-01-30 21:28:05.732 openEarsTest[1700:4107] Setting the variables for the device and starting it.
2014-01-30 21:28:05.733 openEarsTest[1700:4107] Looping through ringbuffer sections and pre-allocating them.
2014-01-30 21:28:05.814 openEarsTest[1700:4107] Started audio output unit.
2014-01-30 21:28:05.816 openEarsTest[1700:4107] Listening.
2014-01-30 21:28:05.818 openEarsTest[1700:60b] Pocketsphinx is now listening.
2014-01-30 21:28:14.579 openEarsTest[1700:4107] Speech detected…
2014-01-30 21:28:14.581 openEarsTest[1700:60b] Pocketsphinx has detected speech.
2014-01-30 21:28:16.694 openEarsTest[1700:4107] Stopping audio unit.
2014-01-30 21:28:16.695 openEarsTest[1700:60b] Pocketsphinx has detected a period of silence, concluding an utterance.
2014-01-30 21:28:16.825 openEarsTest[1700:4107] Audio Output Unit stopped, cleaning up variable states.
2014-01-30 21:28:16.827 openEarsTest[1700:4107] Processing speech, please wait…
INFO: file_omitted(0): cmn_prior_update: from < 61.45 -2.18 -4.31 -5.13 -2.74 -2.52 -0.69 0.55 -0.01 -0.16 -0.04 -0.17 0.06 >
INFO: file_omitted(0): cmn_prior_update: to < 61.84 -1.94 -4.20 -5.15 -2.88 -2.26 -0.75 0.54 -0.02 -0.23 -0.13 -0.27 0.03 >
INFO: file_omitted(0): 1403 words recognized (10/fr)
INFO: file_omitted(0): 6251 senones evaluated (43/fr)
INFO: file_omitted(0): 2000 channels searched (13/fr), 140 1st, 1462 last
INFO: file_omitted(0): 1462 words for which last channels evaluated (10/fr)
INFO: file_omitted(0): 39 candidate words for entering last phone (0/fr)
INFO: file_omitted(0): fwdtree 0.10 CPU 0.068 xRT
INFO: file_omitted(0): fwdtree 2.25 wall 1.561 xRT
INFO: file_omitted(0): Utterance vocabulary contains 1 words
INFO: file_omitted(0): 160 words recognized (1/fr)
INFO: file_omitted(0): 429 senones evaluated (3/fr)
INFO: file_omitted(0): 206 channels searched (1/fr)
INFO: file_omitted(0): 206 words searched (1/fr)
INFO: file_omitted(0): 26 word transitions (0/fr)
INFO: file_omitted(0): fwdflat 0.03 CPU 0.017 xRT
INFO: file_omitted(0): fwdflat 0.03 wall 0.020 xRT
INFO: file_omitted(0): </s> not found in last frame, using <sil>.142 instead
INFO: file_omitted(0): lattice start node <s>.0 end node <sil>.114
INFO: file_omitted(0): Eliminated 0 nodes before end node
INFO: file_omitted(0): Lattice has 18 nodes, 2 links
INFO: file_omitted(0): Normalizer P(O) = alpha(<sil>:114:142) = -1165243
INFO: file_omitted(0): Joint P(O,S) = -1165243 P(S|O) = 0
INFO: file_omitted(0): bestpath 0.00 CPU 0.000 xRT
INFO: file_omitted(0): bestpath 0.00 wall 0.000 xRT
2014-01-30 21:28:16.859 openEarsTest[1700:4107] Pocketsphinx heard “” with a score of (0) and an utterance ID of 000000009.
2014-01-30 21:28:16.860 openEarsTest[1700:4107] Hypothesis was null so we aren’t returning it. If you want null hypotheses to also be returned, set PocketsphinxController’s property returnNullHypotheses to TRUE before starting PocketsphinxController.
2014-01-30 21:28:16.861 openEarsTest[1700:4107] Checking and resetting all audio session settings.
2014-01-30 21:28:16.863 openEarsTest[1700:4107] audioCategory is correct, we will leave it as it is.
2014-01-30 21:28:16.864 openEarsTest[1700:4107] bluetoothInput is correct, we will leave it as it is.
2014-01-30 21:28:16.866 openEarsTest[1700:4107] Output Device: SpeakerAndMicrophone.
2014-01-30 21:28:16.867 openEarsTest[1700:4107] categoryDefaultToSpeaker is correct, we will leave it as it is.
2014-01-30 21:28:16.868 openEarsTest[1700:4107] preferredBufferSize is correct, we will leave it as it is.
2014-01-30 21:28:16.869 openEarsTest[1700:4107] preferredSampleRateCheck is correct, we will leave it as it is.
2014-01-30 21:28:16.870 openEarsTest[1700:4107] Setting the variables for the device and starting it.
2014-01-30 21:28:16.871 openEarsTest[1700:4107] Looping through ringbuffer sections and pre-allocating them.
2014-01-30 21:28:16.951 openEarsTest[1700:4107] Started audio output unit.
2014-01-30 21:28:16.954 openEarsTest[1700:4107] Listening.
2014-01-30 21:28:16.955 openEarsTest[1700:60b] Pocketsphinx is now listening.
2014-01-30 21:28:25.150 openEarsTest[1700:4107] Speech detected…
2014-01-30 21:28:25.152 openEarsTest[1700:60b] Pocketsphinx has detected speech.
2014-01-30 21:28:27.193 openEarsTest[1700:4107] Stopping audio unit.
2014-01-30 21:28:27.193 openEarsTest[1700:60b] Pocketsphinx has detected a period of silence, concluding an utterance.
2014-01-30 21:28:27.322 openEarsTest[1700:4107] Audio Output Unit stopped, cleaning up variable states.
2014-01-30 21:28:27.323 openEarsTest[1700:4107] Processing speech, please wait…
INFO: file_omitted(0): cmn_prior_update: from < 62.18 -1.91 -3.93 -5.23 -2.98 -2.23 -0.79 0.42 0.05 -0.19 -0.03 -0.28 -0.03 >
INFO: file_omitted(0): cmn_prior_update: to < 62.07 -1.85 -4.05 -5.19 -2.94 -2.21 -0.73 0.44 0.01 -0.23 -0.07 -0.27 -0.05 >
INFO: file_omitted(0): 1382 words recognized (10/fr)
INFO: file_omitted(0): 8075 senones evaluated (59/fr)
INFO: file_omitted(0): 3543 channels searched (26/fr), 132 1st, 3039 last
INFO: file_omitted(0): 1461 words for which last channels evaluated (10/fr)
INFO: file_omitted(0): 64 candidate words for entering last phone (0/fr)
INFO: file_omitted(0): fwdtree 0.10 CPU 0.072 xRT
INFO: file_omitted(0): fwdtree 2.17 wall 1.598 xRT
INFO: file_omitted(0): Utterance vocabulary contains 2 words
INFO: file_omitted(0): 232 words recognized (2/fr)
INFO: file_omitted(0): 7906 senones evaluated (58/fr)
INFO: file_omitted(0): 4649 channels searched (34/fr)
INFO: file_omitted(0): 329 words searched (2/fr)
INFO: file_omitted(0): 75 word transitions (0/fr)
INFO: file_omitted(0): fwdflat 0.03 CPU 0.022 xRT
INFO: file_omitted(0): fwdflat 0.04 wall 0.028 xRT
INFO: file_omitted(0): </s> not found in last frame, using OKDEAR.134 instead
INFO: file_omitted(0): lattice start node <s>.0 end node OKDEAR.23
INFO: file_omitted(0): Eliminated 8 nodes before end node
INFO: file_omitted(0): Lattice has 28 nodes, 2 links
INFO: file_omitted(0): Normalizer P(O) = alpha(OKDEAR:23:134) = -1054223
INFO: file_omitted(0): Joint P(O,S) = -1054223 P(S|O) = 0
INFO: file_omitted(0): bestpath 0.00 CPU 0.000 xRT
INFO: file_omitted(0): bestpath 0.00 wall 0.000 xRT
2014-01-30 21:28:27.364 openEarsTest[1700:4107] Pocketsphinx heard “OKDEAR” with a score of (0) and an utterance ID of 000000010.
2014-01-30 21:28:27.366 openEarsTest[1700:4107] Checking and resetting all audio session settings.
2014-01-30 21:28:27.366 openEarsTest[1700:60b] The received hypothesis is OKDEAR with a score of 0 and an ID of 000000010
2014-01-30 21:28:27.369 openEarsTest[1700:4107] audioCategory is correct, we will leave it as it is.
2014-01-30 21:28:27.370 openEarsTest[1700:4107] bluetoothInput is correct, we will leave it as it is.
2014-01-30 21:28:27.371 openEarsTest[1700:4107] Output Device: SpeakerAndMicrophone.
2014-01-30 21:28:27.372 openEarsTest[1700:4107] categoryDefaultToSpeaker is correct, we will leave it as it is.
2014-01-30 21:28:27.373 openEarsTest[1700:4107] preferredBufferSize is correct, we will leave it as it is.
2014-01-30 21:28:27.375 openEarsTest[1700:4107] preferredSampleRateCheck is correct, we will leave it as it is.
2014-01-30 21:28:27.376 openEarsTest[1700:4107] Setting the variables for the device and starting it.
2014-01-30 21:28:27.377 openEarsTest[1700:4107] Looping through ringbuffer sections and pre-allocating them.
2014-01-30 21:28:27.447 openEarsTest[1700:4107] Started audio output unit.
2014-01-30 21:28:27.449 openEarsTest[1700:4107] Listening.
2014-01-30 21:28:27.451 openEarsTest[1700:60b] Pocketsphinx is now listening.
(lldb)It crashes here as I said after about 2 minutes of inactivity:
#0 0x3b07f264 in __commpage_gettimeofday ()
with
EXC_BAD_ACCESS (code=2, address …)I had really light music on the background if it matters.
January 30, 2014 at 8:36 pm #1019996Halle WinklerPolitepixOK, thanks for the info. The silence detection is correct, that just means that it detected enough of a pause to conclude the utterance and start recognition and is normal/required.
b) I have noticed that every time the background noise changed (music on/off) my first three commands went to waste and after that it started recognising. This happens even when I turn the music off so there in no background noise.
This might just be the amount of time needed in order to recover from an abrupt noise level change. The voice audio detection needs a bit of time to adapt and it doesn’t make abrupt changes because that would mean that any idiosyncratic buffer (like a truck driving by) would have a dramatic effect on activity detection.
I also had a crash after two minutes of inactivity.
This is a big deal and I’d appreciate any info you can give me, with OpenEarsLogging and verbosePocketsphinx turned on and then typing in “bt” when you see the blue lldb prompt in the console and showing me the output of that backtrace as well as the complete logging (starting at the beginning) of what OpenEars outputs.
January 30, 2014 at 8:37 pm #1019997Halle WinklerPolitepixWhoops, you slipped past me with all the info I asked for besides the backtrace :) .
January 30, 2014 at 8:38 pm #1019998fdimParticipantI couldn’t confirm issue b. Maybe it was just my idea. I will be doing extensive testing the following days anyway.
January 30, 2014 at 8:45 pm #1020000fdimParticipantOK backtrace coming right up.
A thought/question: Say a user listens to music. Opeanears is calibrated and working properly. And then he pauses music intentionally in order to give a voice command. As you realise there will be an immediate transition to background silence and then the voice will be heard on normal level. How would openears react to that scenario?
January 30, 2014 at 8:49 pm #1020001Halle WinklerPolitepixI’d be interested to know if increasing the number of entries in your vocabulary prevents the crash from occurring.
January 30, 2014 at 8:52 pm #1020002fdimParticipantThis is the backtrace:
2014-01-30 21:48:47.644 openEarsTest[1784:60b] Starting OpenEars logging for OpenEars version 1.65 on 32-bit device: iPad running iOS version: 7.000000
2014-01-30 21:48:47.650 openEarsTest[1784:60b] Normalized array contains the following entries:
(
OKDEAR
)
2014-01-30 21:48:47.680 openEarsTest[1784:60b] Starting dynamic language model generation
2014-01-30 21:48:47.685 openEarsTest[1784:60b] Able to open /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/Library/Caches/NameIWantForMyLanguageModelFiles.corpus for reading
2014-01-30 21:48:47.687 openEarsTest[1784:60b] Able to open /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/Library/Caches/NameIWantForMyLanguageModelFiles_pipe.txt for writing
2014-01-30 21:48:47.688 openEarsTest[1784:60b] Starting text2wfreq_impl
2014-01-30 21:48:47.704 openEarsTest[1784:60b] Done with text2wfreq_impl
2014-01-30 21:48:47.705 openEarsTest[1784:60b] Able to open /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/Library/Caches/NameIWantForMyLanguageModelFiles_pipe.txt for reading.
2014-01-30 21:48:47.707 openEarsTest[1784:60b] Able to open /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/Library/Caches/NameIWantForMyLanguageModelFiles.vocab for reading.
2014-01-30 21:48:47.708 openEarsTest[1784:60b] Starting wfreq2vocab
2014-01-30 21:48:47.710 openEarsTest[1784:60b] Done with wfreq2vocab
2014-01-30 21:48:47.712 openEarsTest[1784:60b] Starting text2idngram
2014-01-30 21:48:47.728 openEarsTest[1784:60b] Done with text2idngram
2014-01-30 21:48:47.732 openEarsTest[1784:60b] Starting idngram2lm2014-01-30 21:48:47.740 openEarsTest[1784:60b] Done with idngram2lm
2014-01-30 21:48:47.741 openEarsTest[1784:60b] Starting sphinx_lm_convert
2014-01-30 21:48:47.748 openEarsTest[1784:60b] Finishing sphinx_lm_convert
2014-01-30 21:48:47.754 openEarsTest[1784:60b] Done creating language model with CMUCLMTK in 0.072978 seconds.
2014-01-30 21:48:47.832 openEarsTest[1784:60b] The word OKDEAR was not found in the dictionary /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/openEarsTest.app/AcousticModelEnglish.bundle/LanguageModelGeneratorLookupList.text/LanguageModelGeneratorLookupList.text.
2014-01-30 21:48:47.833 openEarsTest[1784:60b] Now using the fallback method to look up the word OKDEAR
2014-01-30 21:48:47.834 openEarsTest[1784:60b] 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, or that you are submitting the words in lowercase when they need to be entirely written in uppercase.
2014-01-30 21:48:47.840 openEarsTest[1784:60b] Using convertGraphemes for the word or phrase OKDEAR which doesn’t appear in the dictionary
2014-01-30 21:48:47.865 openEarsTest[1784:60b] I’m done running performDictionaryLookup and it took 0.099134 seconds
2014-01-30 21:48:47.870 openEarsTest[1784:60b] I’m done running dynamic language model generation and it took 0.220802 seconds
2014-01-30 21:48:47.873 openEarsTest[1784:60b] Leaving sample rate at the default of 16000.
2014-01-30 21:48:47.875 openEarsTest[1784:60b] The audio session has never been initialized so we will do that now.
2014-01-30 21:48:47.876 openEarsTest[1784:60b] Checking and resetting all audio session settings.
2014-01-30 21:48:47.878 openEarsTest[1784:60b] audioCategory is incorrect, we will change it.
2014-01-30 21:48:47.882 openEarsTest[1784:60b] audioCategory is now on the correct setting of kAudioSessionCategory_PlayAndRecord.
2014-01-30 21:48:47.883 openEarsTest[1784:60b] bluetoothInput is incorrect, we will change it.
2014-01-30 21:48:47.884 openEarsTest[1784:60b] bluetooth input is now on the correct setting of 1.
2014-01-30 21:48:47.886 openEarsTest[1784:60b] Output Device: SpeakerAndMicrophone.
2014-01-30 21:48:47.887 openEarsTest[1784:60b] categoryDefaultToSpeaker is incorrect, we will change it.
2014-01-30 21:48:47.888 openEarsTest[1784:60b] CategoryDefaultToSpeaker is now on the correct setting of 1.
2014-01-30 21:48:47.890 openEarsTest[1784:60b] preferredBufferSize is incorrect, we will change it.
2014-01-30 21:48:47.891 openEarsTest[1784:60b] PreferredBufferSize is now on the correct setting of 0.128000.
2014-01-30 21:48:47.892 openEarsTest[1784:60b] preferredSampleRateCheck is incorrect, we will change it.
2014-01-30 21:48:47.893 openEarsTest[1784:60b] preferred hardware sample rate is now on the correct setting of 16000.000000.
2014-01-30 21:48:47.919 openEarsTest[1784:60b] AudioSessionManager startAudioSession has reached the end of the initialization.
2014-01-30 21:48:47.922 openEarsTest[1784:60b] Exiting startAudioSession.
2014-01-30 21:48:47.926 openEarsTest[1784:4107] setSecondsOfSilence value of 0.000000 was too large or too small or was NULL, using default of 0.700000.
2014-01-30 21:48:47.930 openEarsTest[1784:4107] Project has these words or phrases in its dictionary:
OKDEAR
2014-01-30 21:48:47.931 openEarsTest[1784:4107] Recognition loop has started
INFO: file_omitted(0): Parsing command line:
\
-lm /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/Library/Caches/NameIWantForMyLanguageModelFiles.DMP \
-dict /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/Library/Caches/NameIWantForMyLanguageModelFiles.dic \
-hmm /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/openEarsTest.app/AcousticModelEnglish.bundle \
-lw 6.500000 \
-samprate 16000Current configuration:
[NAME] [DEFLT] [VALUE]
-agc none none
-agcthresh 2.0 2.000000e+00
-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
-bghist no no
-ceplen 13 13
-cmn current current
-cmninit 8.0 8.0
-compallsen no no
-debug 0
-dict /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/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 /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/openEarsTest.app/AcousticModelEnglish.bundle
-input_endian little little
-jsgf
-kdmaxbbi -1 -1
-kdmaxdepth 0 0
-kdtree
-latsize 5000 5000
-lda
-ldadim 0 0
-lextreedump 0 0
-lifter 0 0
-lm /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/Library/Caches/NameIWantForMyLanguageModelFiles.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
-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-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
-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-022014-01-30 21:48:47.946 openEarsTest[1784:60b] Audio route has changed for the following reason:
2014-01-30 21:48:47.955 openEarsTest[1784:60b] There has been a change of category
2014-01-30 21:48:47.956 openEarsTest[1784:60b] The previous audio route was Speaker
2014-01-30 21:48:47.958 openEarsTest[1784:60b] This is not a case in which OpenEars performs a route change voluntarily. At the close of this function, the audio route is SpeakerAndMicrophone
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 noCurrent 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-02INFO: file_omitted(0): Parsed model-specific feature parameters from /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/openEarsTest.app/AcousticModelEnglish.bundle/feat.params
INFO: file_omitted(0): Initializing feature stream to type: ‘1s_c_d_dd’, ceplen=13, CMN=’current’, VARNORM=’no’, AGC=’none’
INFO: file_omitted(0): mean[0]= 12.00, mean[1..12]= 0.0
INFO: file_omitted(0): Using subvector specification 0-12/13-25/26-38
INFO: file_omitted(0): Reading model definition: /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/openEarsTest.app/AcousticModelEnglish.bundle/mdef
INFO: file_omitted(0): Found byte-order mark BMDF, assuming this is a binary mdef file
INFO: file_omitted(0): Reading binary model definition: /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/openEarsTest.app/AcousticModelEnglish.bundle/mdef
INFO: file_omitted(0): 50 CI-phone, 143047 CD-phone, 3 emitstate/phone, 150 CI-sen, 5150 Sen, 27135 Sen-Seq
INFO: file_omitted(0): Reading HMM transition probability matrices: /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/openEarsTest.app/AcousticModelEnglish.bundle/transition_matrices
INFO: file_omitted(0): Attempting to use SCHMM computation module
INFO: file_omitted(0): Reading mixture gaussian parameter: /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/openEarsTest.app/AcousticModelEnglish.bundle/means
INFO: file_omitted(0): 1 codebook, 3 feature, size:
INFO: file_omitted(0): 256×13
INFO: file_omitted(0): 256×13
INFO: file_omitted(0): 256×13
INFO: file_omitted(0): Reading mixture gaussian parameter: /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/openEarsTest.app/AcousticModelEnglish.bundle/variances
INFO: file_omitted(0): 1 codebook, 3 feature, size:
INFO: file_omitted(0): 256×13
INFO: file_omitted(0): 256×13
INFO: file_omitted(0): 256×13
INFO: file_omitted(0): 0 variance values floored
INFO: file_omitted(0): Loading senones from dump file /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/openEarsTest.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): Maximum top-N: 4 Top-N beams: 0 0 0
INFO: file_omitted(0): Allocating 4108 * 20 bytes (80 KiB) for word entries
INFO: file_omitted(0): Reading main dictionary: /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/Library/Caches/NameIWantForMyLanguageModelFiles.dic
INFO: file_omitted(0): Allocated 0 KiB for strings, 0 KiB for phones
INFO: file_omitted(0): 1 words read
INFO: file_omitted(0): Reading filler dictionary: /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/openEarsTest.app/AcousticModelEnglish.bundle/noisedict
INFO: file_omitted(0): Allocated 0 KiB for strings, 0 KiB for phones
INFO: file_omitted(0): 11 words read
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
INFO: file_omitted(0): No \data\ mark in LM file
INFO: file_omitted(0): Will use memory-mapped I/O for LM file
INFO: file_omitted(0): ngrams 1=3, 2=2, 3=1
INFO: file_omitted(0): 3 = LM.unigrams(+trailer) read
INFO: file_omitted(0): 2 = LM.bigrams(+trailer) read
INFO: file_omitted(0): 1 = LM.trigrams read
INFO: file_omitted(0): 2 = LM.prob2 entries read
INFO: file_omitted(0): 3 = LM.bo_wt2 entries read
INFO: file_omitted(0): 2 = LM.prob3 entries read
INFO: file_omitted(0): 1 = LM.tseg_base entries read
INFO: file_omitted(0): 3 = ascii word strings read
INFO: file_omitted(0): 1 unique initial diphones
INFO: file_omitted(0): 0 root, 0 non-root channels, 12 single-phone words
INFO: file_omitted(0): Creating search tree
INFO: file_omitted(0): before: 0 root, 0 non-root channels, 12 single-phone words
INFO: file_omitted(0): after: max nonroot chan increased to 131
INFO: file_omitted(0): after: 1 root, 3 non-root channels, 11 single-phone words
INFO: file_omitted(0): fwdflat: min_ef_width = 4, max_sf_win = 25
2014-01-30 21:48:48.090 openEarsTest[1784:4107] Starting openAudioDevice on the device.
2014-01-30 21:48:48.091 openEarsTest[1784:4107] Audio unit wrapper successfully created.
2014-01-30 21:48:48.097 openEarsTest[1784:4107] Set audio route to SpeakerAndMicrophone
2014-01-30 21:48:48.099 openEarsTest[1784:4107] There is no CMN plist so we are using the fresh CMN value 47.000000.
2014-01-30 21:48:48.100 openEarsTest[1784:4107] Checking and resetting all audio session settings.
2014-01-30 21:48:48.101 openEarsTest[1784:4107] audioCategory is correct, we will leave it as it is.
2014-01-30 21:48:48.102 openEarsTest[1784:4107] bluetoothInput is correct, we will leave it as it is.
2014-01-30 21:48:48.103 openEarsTest[1784:4107] Output Device: SpeakerAndMicrophone.
2014-01-30 21:48:48.105 openEarsTest[1784:4107] categoryDefaultToSpeaker is correct, we will leave it as it is.
2014-01-30 21:48:48.106 openEarsTest[1784:4107] preferredBufferSize is correct, we will leave it as it is.
2014-01-30 21:48:48.107 openEarsTest[1784:4107] preferredSampleRateCheck is correct, we will leave it as it is.
2014-01-30 21:48:48.107 openEarsTest[1784:4107] Setting the variables for the device and starting it.
2014-01-30 21:48:48.108 openEarsTest[1784:4107] Looping through ringbuffer sections and pre-allocating them.
2014-01-30 21:48:48.585 openEarsTest[1784:4107] Started audio output unit.
2014-01-30 21:48:48.587 openEarsTest[1784:4107] Calibration has started
2014-01-30 21:48:48.587 openEarsTest[1784:60b] Pocketsphinx calibration has started.
2014-01-30 21:48:50.792 openEarsTest[1784:4107] Calibration has completed
2014-01-30 21:48:50.792 openEarsTest[1784:60b] Pocketsphinx calibration is complete.
2014-01-30 21:48:50.795 openEarsTest[1784:4107] Listening.
2014-01-30 21:48:50.796 openEarsTest[1784:60b] Pocketsphinx is now listening.
2014-01-30 21:48:53.097 openEarsTest[1784:4107] Speech detected…
2014-01-30 21:48:53.099 openEarsTest[1784:60b] Pocketsphinx has detected speech.
INFO: file_omitted(0): Resized backpointer table to 10000 entries
2014-01-30 21:48:58.451 openEarsTest[1784:4107] There is reason to suspect the VAD of being out of sync with the current background noise levels in the environment so we will recalibrate.
2014-01-30 21:48:58.453 openEarsTest[1784:4107] Stopping audio unit.
2014-01-30 21:48:58.454 openEarsTest[1784:60b] Pocketsphinx has detected a period of silence, concluding an utterance.
2014-01-30 21:48:58.573 openEarsTest[1784:4107] Audio Output Unit stopped, cleaning up variable states.
2014-01-30 21:48:58.574 openEarsTest[1784:4107] Processing speech, please wait…
INFO: file_omitted(0): cmn_prior_update: from < 47.00 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: file_omitted(0): cmn_prior_update: to < 54.84 -3.25 -3.71 -5.28 -1.54 -2.98 -1.73 0.12 0.03 0.14 0.32 -0.20 0.23 >
INFO: file_omitted(0): 5129 words recognized (9/fr)
INFO: file_omitted(0): 24677 senones evaluated (45/fr)
INFO: file_omitted(0): 7847 channels searched (14/fr), 543 1st, 5732 last
INFO: file_omitted(0): 5472 words for which last channels evaluated (10/fr)
INFO: file_omitted(0): 45 candidate words for entering last phone (0/fr)
INFO: file_omitted(0): fwdtree 0.34 CPU 0.063 xRT
INFO: file_omitted(0): fwdtree 5.48 wall 1.001 xRT
INFO: file_omitted(0): Utterance vocabulary contains 2 words
INFO: file_omitted(0): 595 words recognized (1/fr)
INFO: file_omitted(0): 1638 senones evaluated (3/fr)
INFO: file_omitted(0): 646 channels searched (1/fr)
INFO: file_omitted(0): 646 words searched (1/fr)
INFO: file_omitted(0): 60 word transitions (0/fr)
INFO: file_omitted(0): fwdflat 0.11 CPU 0.020 xRT
INFO: file_omitted(0): fwdflat 0.11 wall 0.020 xRT
INFO: file_omitted(0): lattice start node <s>.0 end node </s>.530
INFO: file_omitted(0): Eliminated 0 nodes before end node
INFO: file_omitted(0): Lattice has 18 nodes, 2 links
INFO: file_omitted(0): Normalizer P(O) = alpha(</s>:530:545) = -4260846
INFO: file_omitted(0): Joint P(O,S) = -4260846 P(S|O) = 0
INFO: file_omitted(0): bestpath 0.00 CPU 0.000 xRT
INFO: file_omitted(0): bestpath 0.00 wall 0.000 xRT
2014-01-30 21:48:58.688 openEarsTest[1784:4107] Pocketsphinx heard “” with a score of (0) and an utterance ID of 000000000.
2014-01-30 21:48:58.689 openEarsTest[1784:4107] Hypothesis was null so we aren’t returning it. If you want null hypotheses to also be returned, set PocketsphinxController’s property returnNullHypotheses to TRUE before starting PocketsphinxController.
2014-01-30 21:48:58.691 openEarsTest[1784:4107] Checking and resetting all audio session settings.
2014-01-30 21:48:58.692 openEarsTest[1784:4107] audioCategory is correct, we will leave it as it is.
2014-01-30 21:48:58.693 openEarsTest[1784:4107] bluetoothInput is correct, we will leave it as it is.
2014-01-30 21:48:58.694 openEarsTest[1784:4107] Output Device: SpeakerAndMicrophone.
2014-01-30 21:48:58.695 openEarsTest[1784:4107] categoryDefaultToSpeaker is correct, we will leave it as it is.
2014-01-30 21:48:58.697 openEarsTest[1784:4107] preferredBufferSize is correct, we will leave it as it is.
2014-01-30 21:48:58.698 openEarsTest[1784:4107] preferredSampleRateCheck is correct, we will leave it as it is.
2014-01-30 21:48:58.699 openEarsTest[1784:4107] Setting the variables for the device and starting it.
2014-01-30 21:48:58.700 openEarsTest[1784:4107] Looping through ringbuffer sections and pre-allocating them.
2014-01-30 21:48:58.825 openEarsTest[1784:4107] Started audio output unit.
2014-01-30 21:48:58.827 openEarsTest[1784:4107] Listening.
2014-01-30 21:48:58.829 openEarsTest[1784:60b] Pocketsphinx is now listening.
2014-01-30 21:48:59.356 openEarsTest[1784:4107] Speech detected…
2014-01-30 21:48:59.358 openEarsTest[1784:60b] Pocketsphinx has detected speech.
INFO: file_omitted(0): Resized backpointer table to 20000 entries
2014-01-30 21:49:09.460 openEarsTest[1784:4107] There is reason to suspect the VAD of being out of sync with the current background noise levels in the environment so we will recalibrate.
2014-01-30 21:49:09.462 openEarsTest[1784:4107] Stopping audio unit.
2014-01-30 21:49:09.462 openEarsTest[1784:60b] Pocketsphinx has detected a period of silence, concluding an utterance.
2014-01-30 21:49:09.580 openEarsTest[1784:4107] Audio Output Unit stopped, cleaning up variable states.
2014-01-30 21:49:09.581 openEarsTest[1784:4107] Processing speech, please wait…
INFO: file_omitted(0): cmn_prior_update: from < 55.50 -3.43 -3.19 -5.34 -2.09 -2.96 -1.57 0.10 0.36 0.11 0.20 -0.27 0.34 >
INFO: file_omitted(0): cmn_prior_update: to < 55.09 -3.32 -3.29 -5.26 -2.20 -2.95 -1.36 0.09 0.44 0.24 0.23 -0.35 0.27 >
INFO: file_omitted(0): 10170 words recognized (10/fr)
INFO: file_omitted(0): 50009 senones evaluated (49/fr)
INFO: file_omitted(0): 17283 channels searched (16/fr), 1025 1st, 13218 last
INFO: file_omitted(0): 10424 words for which last channels evaluated (10/fr)
INFO: file_omitted(0): 368 candidate words for entering last phone (0/fr)
INFO: file_omitted(0): fwdtree 0.67 CPU 0.065 xRT
INFO: file_omitted(0): fwdtree 10.22 wall 0.994 xRT
INFO: file_omitted(0): Utterance vocabulary contains 2 words
INFO: file_omitted(0): 1199 words recognized (1/fr)
INFO: file_omitted(0): 31254 senones evaluated (30/fr)
INFO: file_omitted(0): 17652 channels searched (17/fr)
INFO: file_omitted(0): 1605 words searched (1/fr)
INFO: file_omitted(0): 126 word transitions (0/fr)
INFO: file_omitted(0): fwdflat 0.24 CPU 0.023 xRT
INFO: file_omitted(0): fwdflat 0.24 wall 0.023 xRT
INFO: file_omitted(0): </s> not found in last frame, using <sil>.1027 instead
INFO: file_omitted(0): lattice start node <s>.0 end node <sil>.979
INFO: file_omitted(0): Eliminated 0 nodes before end node
INFO: file_omitted(0): Lattice has 47 nodes, 19 links
INFO: file_omitted(0): Normalizer P(O) = alpha(<sil>:979:1027) = -7500114
INFO: file_omitted(0): Joint P(O,S) = -7500114 P(S|O) = 0
INFO: file_omitted(0): bestpath 0.00 CPU 0.000 xRT
INFO: file_omitted(0): bestpath 0.00 wall 0.000 xRT
2014-01-30 21:49:09.820 openEarsTest[1784:4107] Pocketsphinx heard “OKDEAR” with a score of (0) and an utterance ID of 000000001.
2014-01-30 21:49:09.821 openEarsTest[1784:60b] The received hypothesis is OKDEAR with a score of 0 and an ID of 000000001
2014-01-30 21:49:09.822 openEarsTest[1784:4107] Checking and resetting all audio session settings.
2014-01-30 21:49:09.824 openEarsTest[1784:4107] audioCategory is correct, we will leave it as it is.
2014-01-30 21:49:09.825 openEarsTest[1784:4107] bluetoothInput is correct, we will leave it as it is.
2014-01-30 21:49:09.826 openEarsTest[1784:4107] Output Device: SpeakerAndMicrophone.
2014-01-30 21:49:09.827 openEarsTest[1784:4107] categoryDefaultToSpeaker is correct, we will leave it as it is.
2014-01-30 21:49:09.828 openEarsTest[1784:4107] preferredBufferSize is correct, we will leave it as it is.
2014-01-30 21:49:09.829 openEarsTest[1784:4107] preferredSampleRateCheck is correct, we will leave it as it is.
2014-01-30 21:49:09.830 openEarsTest[1784:4107] Setting the variables for the device and starting it.
2014-01-30 21:49:09.831 openEarsTest[1784:4107] Looping through ringbuffer sections and pre-allocating them.
2014-01-30 21:49:09.960 openEarsTest[1784:4107] Started audio output unit.
2014-01-30 21:49:09.963 openEarsTest[1784:4107] Listening.
2014-01-30 21:49:09.965 openEarsTest[1784:60b] Pocketsphinx is now listening.
2014-01-30 21:49:10.493 openEarsTest[1784:4107] Speech detected…
2014-01-30 21:49:10.495 openEarsTest[1784:60b] Pocketsphinx has detected speech.
2014-01-30 21:49:11.755 openEarsTest[1784:4107] Stopping audio unit.
2014-01-30 21:49:11.755 openEarsTest[1784:60b] Pocketsphinx has detected a period of silence, concluding an utterance.
2014-01-30 21:49:11.882 openEarsTest[1784:4107] Audio Output Unit stopped, cleaning up variable states.
2014-01-30 21:49:11.885 openEarsTest[1784:4107] Processing speech, please wait…
INFO: file_omitted(0): cmn_prior_update: from < 55.09 -3.32 -3.29 -5.26 -2.20 -2.95 -1.36 0.09 0.44 0.24 0.23 -0.35 0.27 >
INFO: file_omitted(0): cmn_prior_update: to < 54.65 -3.51 -3.15 -5.31 -2.11 -2.99 -1.34 0.08 0.47 0.23 0.22 -0.35 0.26 >
INFO: file_omitted(0): 612 words recognized (9/fr)
INFO: file_omitted(0): 2772 senones evaluated (41/fr)
INFO: file_omitted(0): 879 channels searched (13/fr), 63 1st, 663 last
INFO: file_omitted(0): 663 words for which last channels evaluated (9/fr)
INFO: file_omitted(0): 1 candidate words for entering last phone (0/fr)
INFO: file_omitted(0): fwdtree 0.05 CPU 0.074 xRT
INFO: file_omitted(0): fwdtree 1.39 wall 2.078 xRT
INFO: file_omitted(0): Utterance vocabulary contains 2 words
INFO: file_omitted(0): 117 words recognized (2/fr)
INFO: file_omitted(0): 198 senones evaluated (3/fr)
INFO: file_omitted(0): 172 channels searched (2/fr)
INFO: file_omitted(0): 172 words searched (2/fr)
INFO: file_omitted(0): 64 word transitions (0/fr)
INFO: file_omitted(0): fwdflat 0.01 CPU 0.015 xRT
INFO: file_omitted(0): fwdflat 0.01 wall 0.017 xRT
INFO: file_omitted(0): lattice start node <s>.0 end node </s>.50
INFO: file_omitted(0): Eliminated 0 nodes before end node
INFO: file_omitted(0): Lattice has 19 nodes, 5 links
INFO: file_omitted(0): Normalizer P(O) = alpha(</s>:50:65) = -487860
INFO: file_omitted(0): Joint P(O,S) = -488123 P(S|O) = -263
INFO: file_omitted(0): bestpath 0.00 CPU 0.000 xRT
INFO: file_omitted(0): bestpath 0.00 wall 0.000 xRT
2014-01-30 21:49:11.899 openEarsTest[1784:4107] Pocketsphinx heard “” with a score of (-263) and an utterance ID of 000000002.
2014-01-30 21:49:11.900 openEarsTest[1784:4107] Hypothesis was null so we aren’t returning it. If you want null hypotheses to also be returned, set PocketsphinxController’s property returnNullHypotheses to TRUE before starting PocketsphinxController.
2014-01-30 21:49:11.902 openEarsTest[1784:4107] Checking and resetting all audio session settings.
2014-01-30 21:49:11.904 openEarsTest[1784:4107] audioCategory is correct, we will leave it as it is.
2014-01-30 21:49:11.905 openEarsTest[1784:4107] bluetoothInput is correct, we will leave it as it is.
2014-01-30 21:49:11.906 openEarsTest[1784:4107] Output Device: SpeakerAndMicrophone.
2014-01-30 21:49:11.907 openEarsTest[1784:4107] categoryDefaultToSpeaker is correct, we will leave it as it is.
2014-01-30 21:49:11.908 openEarsTest[1784:4107] preferredBufferSize is correct, we will leave it as it is.
2014-01-30 21:49:11.910 openEarsTest[1784:4107] preferredSampleRateCheck is correct, we will leave it as it is.
2014-01-30 21:49:11.911 openEarsTest[1784:4107] Setting the variables for the device and starting it.
2014-01-30 21:49:11.912 openEarsTest[1784:4107] Looping through ringbuffer sections and pre-allocating them.
2014-01-30 21:49:12.009 openEarsTest[1784:4107] Started audio output unit.
2014-01-30 21:49:12.012 openEarsTest[1784:4107] Listening.
2014-01-30 21:49:12.013 openEarsTest[1784:60b] Pocketsphinx is now listening.
(lldb) bt
* thread #5: tid = 0x55600, 0x3b07f264 libsystem_kernel.dylib`__commpage_gettimeofday + 4, stop reason = EXC_BAD_ACCESS (code=2, address=0x2a98ff8)
frame #0: 0x3b07f264 libsystem_kernel.dylib`__commpage_gettimeofday + 4
frame #1: 0x3afea48a libsystem_c.dylib`gettimeofday + 26
frame #2: 0x30251176 CoreFoundation`CFAbsoluteTimeGetCurrent + 14
frame #3: 0x0014ceb2 openEarsTest`readBufferContents + 386
frame #4: 0x0014a154 openEarsTest`cont_ad_read_internal + 184
frame #5: 0x00149cc8 openEarsTest`cont_ad_read + 200
frame #6: 0x0015065e openEarsTest`-[ContinuousModel listeningForSpeechWithSleepTime:withSpeechData:andBuffer:] + 482
[…] identical frames 7-3472 removed.
frame #3473: 0x0015071a openEarsTest`-[ContinuousModel listeningForSpeechWithSleepTime:withSpeechData:andBuffer:] + 670
frame #3474: 0x00152898 openEarsTest`-[ContinuousModel listeningLoopWithLanguageModelAtPath:dictionaryAtPath:acousticModelAtPath:languageModelIsJSGF:] + 1196
frame #3475: 0x0015c564 openEarsTest`-[PocketsphinxController startVoiceRecognitionThreadAutoreleasePoolWithArray:] + 1484
frame #3476: 0x30d01c36 Foundation`__NSThread__main__ + 1062
frame #3477: 0x3b0e8c5c libsystem_pthread.dylib`_pthread_body + 140
frame #3478: 0x3b0e8bce libsystem_pthread.dylib`_pthread_start + 102
(lldb)January 30, 2014 at 8:57 pm #1020003fdimParticipantAnd this is the crash log with more words in the dictionary. Sorry :)
2014-01-30 21:53:55.154 openEarsTest[1800:60b] Starting OpenEars logging for OpenEars version 1.65 on 32-bit device: iPad running iOS version: 7.000000
2014-01-30 21:53:55.160 openEarsTest[1800:60b] Normalized array contains the following entries:
(
OKDEAR,
OKRAINDEER,
OKJOE,
OKCRASH
)
2014-01-30 21:53:55.206 openEarsTest[1800:60b] Starting dynamic language model generation
2014-01-30 21:53:55.211 openEarsTest[1800:60b] Able to open /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/Library/Caches/NameIWantForMyLanguageModelFiles.corpus for reading
2014-01-30 21:53:55.214 openEarsTest[1800:60b] Able to open /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/Library/Caches/NameIWantForMyLanguageModelFiles_pipe.txt for writing
2014-01-30 21:53:55.215 openEarsTest[1800:60b] Starting text2wfreq_impl
2014-01-30 21:53:55.228 openEarsTest[1800:60b] Done with text2wfreq_impl
2014-01-30 21:53:55.230 openEarsTest[1800:60b] Able to open /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/Library/Caches/NameIWantForMyLanguageModelFiles_pipe.txt for reading.
2014-01-30 21:53:55.232 openEarsTest[1800:60b] Able to open /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/Library/Caches/NameIWantForMyLanguageModelFiles.vocab for reading.
2014-01-30 21:53:55.233 openEarsTest[1800:60b] Starting wfreq2vocab
2014-01-30 21:53:55.235 openEarsTest[1800:60b] Done with wfreq2vocab
2014-01-30 21:53:55.236 openEarsTest[1800:60b] Starting text2idngram
2014-01-30 21:53:55.253 openEarsTest[1800:60b] Done with text2idngram
2014-01-30 21:53:55.257 openEarsTest[1800:60b] Starting idngram2lm2014-01-30 21:53:55.265 openEarsTest[1800:60b] Done with idngram2lm
2014-01-30 21:53:55.266 openEarsTest[1800:60b] Starting sphinx_lm_convert
2014-01-30 21:53:55.273 openEarsTest[1800:60b] Finishing sphinx_lm_convert
2014-01-30 21:53:55.279 openEarsTest[1800:60b] Done creating language model with CMUCLMTK in 0.071863 seconds.
2014-01-30 21:53:55.373 openEarsTest[1800:60b] The word OKCRASH was not found in the dictionary /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/openEarsTest.app/AcousticModelEnglish.bundle/LanguageModelGeneratorLookupList.text/LanguageModelGeneratorLookupList.text.
2014-01-30 21:53:55.374 openEarsTest[1800:60b] Now using the fallback method to look up the word OKCRASH
2014-01-30 21:53:55.375 openEarsTest[1800:60b] 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, or that you are submitting the words in lowercase when they need to be entirely written in uppercase.
2014-01-30 21:53:55.380 openEarsTest[1800:60b] Using convertGraphemes for the word or phrase OKCRASH which doesn’t appear in the dictionary
2014-01-30 21:53:55.481 openEarsTest[1800:60b] The word OKDEAR was not found in the dictionary /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/openEarsTest.app/AcousticModelEnglish.bundle/LanguageModelGeneratorLookupList.text/LanguageModelGeneratorLookupList.text.
2014-01-30 21:53:55.482 openEarsTest[1800:60b] Now using the fallback method to look up the word OKDEAR
2014-01-30 21:53:55.483 openEarsTest[1800:60b] 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, or that you are submitting the words in lowercase when they need to be entirely written in uppercase.
2014-01-30 21:53:55.484 openEarsTest[1800:60b] Using convertGraphemes for the word or phrase OKDEAR which doesn’t appear in the dictionary
2014-01-30 21:53:55.569 openEarsTest[1800:60b] The word OKJOE was not found in the dictionary /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/openEarsTest.app/AcousticModelEnglish.bundle/LanguageModelGeneratorLookupList.text/LanguageModelGeneratorLookupList.text.
2014-01-30 21:53:55.570 openEarsTest[1800:60b] Now using the fallback method to look up the word OKJOE
2014-01-30 21:53:55.571 openEarsTest[1800:60b] 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, or that you are submitting the words in lowercase when they need to be entirely written in uppercase.
2014-01-30 21:53:55.573 openEarsTest[1800:60b] Using convertGraphemes for the word or phrase OKJOE which doesn’t appear in the dictionary
2014-01-30 21:53:55.649 openEarsTest[1800:60b] The word OKRAINDEER was not found in the dictionary /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/openEarsTest.app/AcousticModelEnglish.bundle/LanguageModelGeneratorLookupList.text/LanguageModelGeneratorLookupList.text.
2014-01-30 21:53:55.650 openEarsTest[1800:60b] Now using the fallback method to look up the word OKRAINDEER
2014-01-30 21:53:55.651 openEarsTest[1800:60b] 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, or that you are submitting the words in lowercase when they need to be entirely written in uppercase.
2014-01-30 21:53:55.652 openEarsTest[1800:60b] Using convertGraphemes for the word or phrase OKRAINDEER which doesn’t appear in the dictionary
2014-01-30 21:53:55.662 openEarsTest[1800:60b] I’m done running performDictionaryLookup and it took 0.359341 seconds
2014-01-30 21:53:55.667 openEarsTest[1800:60b] I’m done running dynamic language model generation and it took 0.507939 seconds
2014-01-30 21:53:55.670 openEarsTest[1800:60b] Leaving sample rate at the default of 16000.
2014-01-30 21:53:55.672 openEarsTest[1800:60b] The audio session has never been initialized so we will do that now.
2014-01-30 21:53:55.673 openEarsTest[1800:60b] Checking and resetting all audio session settings.
2014-01-30 21:53:55.675 openEarsTest[1800:60b] audioCategory is incorrect, we will change it.
2014-01-30 21:53:55.679 openEarsTest[1800:60b] audioCategory is now on the correct setting of kAudioSessionCategory_PlayAndRecord.
2014-01-30 21:53:55.680 openEarsTest[1800:60b] bluetoothInput is incorrect, we will change it.
2014-01-30 21:53:55.681 openEarsTest[1800:60b] bluetooth input is now on the correct setting of 1.
2014-01-30 21:53:55.683 openEarsTest[1800:60b] Output Device: SpeakerAndMicrophone.
2014-01-30 21:53:55.684 openEarsTest[1800:60b] categoryDefaultToSpeaker is incorrect, we will change it.
2014-01-30 21:53:55.685 openEarsTest[1800:60b] CategoryDefaultToSpeaker is now on the correct setting of 1.
2014-01-30 21:53:55.687 openEarsTest[1800:60b] preferredBufferSize is incorrect, we will change it.
2014-01-30 21:53:55.688 openEarsTest[1800:60b] PreferredBufferSize is now on the correct setting of 0.128000.
2014-01-30 21:53:55.689 openEarsTest[1800:60b] preferredSampleRateCheck is incorrect, we will change it.
2014-01-30 21:53:55.690 openEarsTest[1800:60b] preferred hardware sample rate is now on the correct setting of 16000.000000.
2014-01-30 21:53:55.717 openEarsTest[1800:60b] AudioSessionManager startAudioSession has reached the end of the initialization.
2014-01-30 21:53:55.719 openEarsTest[1800:60b] Exiting startAudioSession.
2014-01-30 21:53:55.723 openEarsTest[1800:4107] setSecondsOfSilence value of 0.000000 was too large or too small or was NULL, using default of 0.700000.
2014-01-30 21:53:55.731 openEarsTest[1800:4107] Project has these words or phrases in its dictionary:
OKCRASH
OKDEAR
OKJOE
OKRAINDEER
2014-01-30 21:53:55.734 openEarsTest[1800:4107] Recognition loop has started
INFO: file_omitted(0): Parsing command line:
\
-lm /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/Library/Caches/NameIWantForMyLanguageModelFiles.DMP \
-dict /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/Library/Caches/NameIWantForMyLanguageModelFiles.dic \
-hmm /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/openEarsTest.app/AcousticModelEnglish.bundle \
-lw 6.500000 \
-samprate 16000Current configuration:
[NAME] [DEFLT] [VALUE]
-agc none none
-agcthresh 2.0 2.000000e+00
-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
-bghist no no
-ceplen 13 13
-cmn current current
-cmninit 8.0 8.0
-compallsen no no
-debug 0
-dict /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/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 /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/openEarsTest.app/AcousticModelEnglish.bundle
-input_endian little little
-jsgf
-kdmaxbbi -1 -1
-kdmaxdepth 0 0
-kdtree
-latsize 5000 5000
-lda
-ldadim 0 0
-lextreedump 0 0
-lifter 0 0
-lm /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/Library/Caches/NameIWantForMyLanguageModelFiles.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
-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-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
-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
2014-01-30 21:53:55.737 openEarsTest[1800:60b] Audio route has changed for the following reason:
2014-01-30 21:53:55.753 openEarsTest[1800:60b] There has been a change of category
2014-01-30 21:53:55.756 openEarsTest[1800:60b] The previous audio route was Speaker
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 noCurrent 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 1002014-01-30 21:53:55.771 openEarsTest[1800:60b] This is not a case in which OpenEars performs a route change voluntarily. At the close of this function, the audio route is SpeakerAndMicrophone-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-02INFO: file_omitted(0): Parsed model-specific feature parameters from /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/openEarsTest.app/AcousticModelEnglish.bundle/feat.params
INFO: file_omitted(0): Initializing feature stream to type: ‘1s_c_d_dd’, ceplen=13, CMN=’current’, VARNORM=’no’, AGC=’none’
INFO: file_omitted(0): mean[0]= 12.00, mean[1..12]= 0.0
INFO: file_omitted(0): Using subvector specification 0-12/13-25/26-38
INFO: file_omitted(0): Reading model definition: /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/openEarsTest.app/AcousticModelEnglish.bundle/mdef
INFO: file_omitted(0): Found byte-order mark BMDF, assuming this is a binary mdef file
INFO: file_omitted(0): Reading binary model definition: /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/openEarsTest.app/AcousticModelEnglish.bundle/mdef
INFO: file_omitted(0): 50 CI-phone, 143047 CD-phone, 3 emitstate/phone, 150 CI-sen, 5150 Sen, 27135 Sen-Seq
INFO: file_omitted(0): Reading HMM transition probability matrices: /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/openEarsTest.app/AcousticModelEnglish.bundle/transition_matrices
INFO: file_omitted(0): Attempting to use SCHMM computation module
INFO: file_omitted(0): Reading mixture gaussian parameter: /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/openEarsTest.app/AcousticModelEnglish.bundle/means
INFO: file_omitted(0): 1 codebook, 3 feature, size:
INFO: file_omitted(0): 256×13
INFO: file_omitted(0): 256×13
INFO: file_omitted(0): 256×13
INFO: file_omitted(0): Reading mixture gaussian parameter: /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/openEarsTest.app/AcousticModelEnglish.bundle/variances
INFO: file_omitted(0): 1 codebook, 3 feature, size:
INFO: file_omitted(0): 256×13
INFO: file_omitted(0): 256×13
INFO: file_omitted(0): 256×13
INFO: file_omitted(0): 0 variance values floored
INFO: file_omitted(0): Loading senones from dump file /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/openEarsTest.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): Maximum top-N: 4 Top-N beams: 0 0 0
INFO: file_omitted(0): Allocating 4111 * 20 bytes (80 KiB) for word entries
INFO: file_omitted(0): Reading main dictionary: /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/Library/Caches/NameIWantForMyLanguageModelFiles.dic
INFO: file_omitted(0): Allocated 0 KiB for strings, 0 KiB for phones
INFO: file_omitted(0): 4 words read
INFO: file_omitted(0): Reading filler dictionary: /var/mobile/Applications/13E4E949-1A48-411D-8F43-502908075A59/openEarsTest.app/AcousticModelEnglish.bundle/noisedict
INFO: file_omitted(0): Allocated 0 KiB for strings, 0 KiB for phones
INFO: file_omitted(0): 11 words read
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
INFO: file_omitted(0): No \data\ mark in LM file
INFO: file_omitted(0): Will use memory-mapped I/O for LM file
INFO: file_omitted(0): ngrams 1=6, 2=8, 3=4
INFO: file_omitted(0): 6 = LM.unigrams(+trailer) read
INFO: file_omitted(0): 8 = LM.bigrams(+trailer) read
INFO: file_omitted(0): 4 = LM.trigrams read
INFO: file_omitted(0): 3 = LM.prob2 entries read
INFO: file_omitted(0): 3 = LM.bo_wt2 entries read
INFO: file_omitted(0): 2 = LM.prob3 entries read
INFO: file_omitted(0): 1 = LM.tseg_base entries read
INFO: file_omitted(0): 6 = ascii word strings read
INFO: file_omitted(0): 2 unique initial diphones
INFO: file_omitted(0): 0 root, 0 non-root channels, 12 single-phone words
INFO: file_omitted(0): Creating search tree
INFO: file_omitted(0): before: 0 root, 0 non-root channels, 12 single-phone words
INFO: file_omitted(0): after: max nonroot chan increased to 141
INFO: file_omitted(0): after: 2 root, 13 non-root channels, 11 single-phone words
INFO: file_omitted(0): fwdflat: min_ef_width = 4, max_sf_win = 25
2014-01-30 21:53:55.906 openEarsTest[1800:4107] Starting openAudioDevice on the device.
2014-01-30 21:53:55.907 openEarsTest[1800:4107] Audio unit wrapper successfully created.
2014-01-30 21:53:55.915 openEarsTest[1800:4107] Set audio route to SpeakerAndMicrophone
2014-01-30 21:53:55.917 openEarsTest[1800:4107] There is no CMN plist so we are using the fresh CMN value 47.000000.
2014-01-30 21:53:55.918 openEarsTest[1800:4107] Checking and resetting all audio session settings.
2014-01-30 21:53:55.919 openEarsTest[1800:4107] audioCategory is correct, we will leave it as it is.
2014-01-30 21:53:55.921 openEarsTest[1800:4107] bluetoothInput is correct, we will leave it as it is.
2014-01-30 21:53:55.922 openEarsTest[1800:4107] Output Device: SpeakerAndMicrophone.
2014-01-30 21:53:55.923 openEarsTest[1800:4107] categoryDefaultToSpeaker is correct, we will leave it as it is.
2014-01-30 21:53:55.924 openEarsTest[1800:4107] preferredBufferSize is correct, we will leave it as it is.
2014-01-30 21:53:55.926 openEarsTest[1800:4107] preferredSampleRateCheck is correct, we will leave it as it is.
2014-01-30 21:53:55.927 openEarsTest[1800:4107] Setting the variables for the device and starting it.
2014-01-30 21:53:55.928 openEarsTest[1800:4107] Looping through ringbuffer sections and pre-allocating them.
2014-01-30 21:53:56.408 openEarsTest[1800:4107] Started audio output unit.
2014-01-30 21:53:56.410 openEarsTest[1800:4107] Calibration has started
2014-01-30 21:53:56.410 openEarsTest[1800:60b] Pocketsphinx calibration has started.
2014-01-30 21:53:58.615 openEarsTest[1800:4107] Calibration has completed
2014-01-30 21:53:58.615 openEarsTest[1800:60b] Pocketsphinx calibration is complete.
2014-01-30 21:53:58.617 openEarsTest[1800:4107] Listening.
2014-01-30 21:53:58.619 openEarsTest[1800:60b] Pocketsphinx is now listening.
(lldb) bt
* thread #5: tid = 0x55a56, 0x00127d18 openEarsTest`readBufferContents, stop reason = EXC_BAD_ACCESS (code=2, address=0x2a73fec)
frame #0: 0x00127d18 openEarsTest`readBufferContents
frame #1: 0x0012513c openEarsTest`cont_ad_read_internal + 184
frame #2: 0x00124cb0 openEarsTest`cont_ad_read + 200
frame #3: 0x0012b646 openEarsTest`-[ContinuousModel listeningForSpeechWithSleepTime:withSpeechData:andBuffer:] + 482
[….] identical frames 4-3470 removed.
frame #3471: 0x0012b702 openEarsTest`-[ContinuousModel listeningForSpeechWithSleepTime:withSpeechData:andBuffer:] + 670
frame #3472: 0x0012d880 openEarsTest`-[ContinuousModel listeningLoopWithLanguageModelAtPath:dictionaryAtPath:acousticModelAtPath:languageModelIsJSGF:] + 1196
frame #3473: 0x0013754c openEarsTest`-[PocketsphinxController startVoiceRecognitionThreadAutoreleasePoolWithArray:] + 1484
frame #3474: 0x30d01c36 Foundation`__NSThread__main__ + 1062
frame #3475: 0x3b0e8c5c libsystem_pthread.dylib`_pthread_body + 140
frame #3476: 0x3b0e8bce libsystem_pthread.dylib`_pthread_start + 102
(lldb)January 30, 2014 at 8:59 pm #1020004Halle WinklerPolitepixMy last reply was referring to your reply before your last one.
A thought/question: Say a user listens to music. Opeanears is calibrated and working properly. And then he pauses music intentionally in order to give a voice command. As you realise there will be an immediate transition to background silence and then the voice will be heard on normal level. How would openears react to that scenario?
OpenEars isn’t designed to listen for speech with the internal mic while the same device is being used to play sound out the built-in speaker in an open environment, which is just a worst-case recognition scenario unfortunately. The design is that you suspend listening while playing music or other sound from the device if you are using the speaker and internal mic together. So in this scenario, you would want to use it according to its design and have it not listening while the music is on, and to resume listening when the user pauses the music to speak.
January 31, 2014 at 4:49 pm #1020010Halle WinklerPolitepixOK, there is a new beta fixing the silence issue, check it out and let me know if it’s working for you:
https://www.politepix.com/wp-content/uploads/OpenEarsDistributionBeta.tar.bz2
February 1, 2014 at 10:49 am #1020012fdimParticipantOK I believe the crash is fixed now.
OpenEars isn’t designed to listen for speech with the internal mic while the same device is being used to play sound out the built-in speaker in an open environment, which is just a worst-case recognition scenario unfortunately. The design is that you suspend listening while playing music or other sound from the device if you are using the speaker and internal mic together. So in this scenario, you would want to use it according to its design and have it not listening while the music is on, and to resume listening when the user pauses the music to speak.
Is there any specific reason for this? Is that because openears takes over the audio session or because of something else? Actually we have run some tests in the past, setting a mixed session and recognition worked even with music on (at least on low volume). We were using an audio out thought sending music to external speakers.
February 1, 2014 at 11:13 am #1020013Halle WinklerPolitepixI’m glad to hear it. The only reason it isn’t designed for that and is designed for the use of suspend/resume during internal speaker output is just because it doesn’t work well to do speech recognition using a mic that is right next to a speaker from which continuous sound is emitting. But your case with an audio out is a different situation that can work better, if the audio out is leading to speakers that are not next to the mic and not closer to the mic than the user is, and isn’t operating with similar volume at distance to user speech. But it does have to be notably quieter at the mic input or it will cause a lot of recognition confusion.
So in that case the answer to your question above would be that the calibration will adapt to the decrease in background noise. The case you were seeing in which there was no adaptation was because of a stuck value in the VAD (the only circumstance I’m aware of this happening currently is when there is a fast, persistent and very large increase in background noise levels, but in the cases in which this happens there will now be a quick recalibration as soon as the “stuckness” gets on the radar after a few seconds). Otherwise the expected behavior during increases or decreases of background noise levels is adaptation, although the adaptation will not be instantaneous.
-
AuthorPosts
- You must be logged in to reply to this topic.