Menu

PocketSphinx decoding error

Help
mehmet
2011-09-13
2012-09-22
  • mehmet

    mehmet - 2011-09-13

    Hi,
    I am using pocketsphinx to decode Arabic audio files and extract the
    corresponding transcriptions. Time to time, audio files to be decoded may be
    too large to be decoded by PocketSphix, and I had to split them by using
    mp3splt toolkit. Hence, I divide them down to 1 or 2 min length audio segments
    if they are longer than it. However, PocketSphinx chrashes at some audio
    segments. The crash output is following:

    INFO: acmod.c(242): Parsed model-specific feature parameters from
    /home/mehmet/sphinx/am/ar-ipsos/feat.params
    INFO: fe_interface.c(289): You are using the internal mechanism to generate
    the seed.
    INFO: feat.c(684): Initializing feature stream to type: '1s_c_d_dd',
    ceplen=13, CMN='current', VARNORM='no', AGC='none'
    INFO: cmn.c(142): mean= 12.00, mean= 0.0
    INFO: mdef.c(520): Reading model definition: /home/mehmet/sphinx/am/ar-
    ipsos/mdef
    INFO: bin_mdef.c(173): Allocating 62086 * 8 bytes (485 KiB) for CD tree
    INFO: tmat.c(205): Reading HMM transition probability matrices:
    /home/mehmet/sphinx/am/ar-ipsos/transition_matrices
    INFO: acmod.c(117): Attempting to use SCHMM computation module
    INFO: ms_gauden.c(198): Reading mixture gaussian parameter:
    /home/mehmet/sphinx/am/ar-ipsos/means
    INFO: ms_gauden.c(292): 6105 codebook, 1 feature, size:
    INFO: ms_gauden.c(294): 16x39
    INFO: ms_gauden.c(198): Reading mixture gaussian parameter:
    /home/mehmet/sphinx/am/ar-ipsos/variances
    INFO: ms_gauden.c(292): 6105 codebook, 1 feature, size:
    INFO: ms_gauden.c(294): 16x39
    INFO: ms_gauden.c(354): 92 variance values floored
    INFO: acmod.c(119): Attempting to use PTHMM computation module
    INFO: ms_gauden.c(198): Reading mixture gaussian parameter:
    /home/mehmet/sphinx/am/ar-ipsos/means
    INFO: ms_gauden.c(292): 6105 codebook, 1 feature, size:
    INFO: ms_gauden.c(294): 16x39
    INFO: ms_gauden.c(198): Reading mixture gaussian parameter:
    /home/mehmet/sphinx/am/ar-ipsos/variances
    INFO: ms_gauden.c(292): 6105 codebook, 1 feature, size:
    INFO: ms_gauden.c(294): 16x39
    INFO: ms_gauden.c(354): 92 variance values floored
    INFO: ptm_mgau.c(800): Number of codebooks exceeds 256: 6105
    INFO: acmod.c(121): Falling back to general multi-stream GMM computation
    INFO: ms_gauden.c(198): Reading mixture gaussian parameter:
    /home/mehmet/sphinx/am/ar-ipsos/means
    INFO: ms_gauden.c(292): 6105 codebook, 1 feature, size:
    INFO: ms_gauden.c(294): 16x39
    INFO: ms_gauden.c(198): Reading mixture gaussian parameter:
    /home/mehmet/sphinx/am/ar-ipsos/variances
    INFO: ms_gauden.c(292): 6105 codebook, 1 feature, size:
    INFO: ms_gauden.c(294): 16x39
    INFO: ms_gauden.c(354): 92 variance values floored
    INFO: ms_senone.c(160): Reading senone mixture weights: /home/mehmet/sphinx/am
    /ar-ipsos/mixture_weights
    INFO: ms_senone.c(211): Truncating senone logs3(pdf) values by 10 bits
    INFO: ms_senone.c(218): Not transposing mixture weights in memory
    INFO: ms_senone.c(277): Read mixture weights for 6105 senones: 1 features x 16
    codewords
    INFO: ms_senone.c(331): Mapping senones to individual codebooks
    INFO: ms_mgau.c(122): The value of topn: 4
    INFO: dict.c(306): Allocating 111223 * 32 bytes (3475 KiB) for word entries
    INFO: dict.c(321): Reading main dictionary: /home/mehmet/sphinx/lm/ar-
    ipsos/arabic.dic
    INFO: dict.c(212): Allocated 983 KiB for strings, 2022 KiB for phones
    INFO: dict.c(324): 107124 words read
    INFO: dict.c(330): Reading filler dictionary: /home/mehmet/sphinx/am/ar-
    ipsos/noisedict
    INFO: dict.c(212): Allocated 0 KiB for strings, 0 KiB for phones
    INFO: dict.c(333): 3 words read
    INFO: dict2pid.c(396): Building PID tables for dictionary
    INFO: dict2pid.c(404): Allocating 35^3 * 2 bytes (83 KiB) for word-initial
    triphones
    INFO: dict2pid.c(131): Allocated 29680 bytes (28 KiB) for word-final triphones
    INFO: dict2pid.c(195): Allocated 29680 bytes (28 KiB) for single-phone word
    triphones
    INFO: ngram_model_arpa.c(77): No \data\ mark in LM file
    INFO: ngram_model_dmp.c(142): Will use memory-mapped I/O for LM file
    INFO: ngram_model_dmp.c(196): ngrams 1=65530, 2=564123, 3=867799
    INFO: ngram_model_dmp.c(242): 65530 = LM.unigrams(+trailer) read
    INFO: ngram_model_dmp.c(291): 564123 = LM.bigrams(+trailer) read
    INFO: ngram_model_dmp.c(317): 867799 = LM.trigrams read
    INFO: ngram_model_dmp.c(342): 8396 = LM.prob2 entries read
    INFO: ngram_model_dmp.c(362): 8781 = LM.bo_wt2 entries read
    INFO: ngram_model_dmp.c(382): 3170 = LM.prob3 entries read
    INFO: ngram_model_dmp.c(410): 1102 = LM.tseg_base entries read
    INFO: ngram_model_dmp.c(466): 65530 = ascii word strings read
    INFO: ngram_search_fwdtree.c(99): 543 unique initial diphones
    INFO: ngram_search_fwdtree.c(147): 0 root, 0 non-root channels, 14 single-
    phone words
    INFO: ngram_search_fwdtree.c(186): Creating search tree
    INFO: ngram_search_fwdtree.c(191): before: 0 root, 0 non-root channels, 14
    single-phone words
    INFO: ngram_search_fwdtree.c(326): after: max nonroot chan increased to 235750
    INFO: ngram_search_fwdtree.c(338): after: 454 root, 235622 non-root channels,
    12 single-phone words
    INFO: ngram_search_fwdflat.c(156): fwdflat: min_ef_width = 4, max_sf_win = 25
    INFO: cmn.c(175): CMN: 10.95 0.07 -0.34 -0.10 -0.38 -0.31 -0.18 -0.36 -0.16
    -0.13 -0.06 -0.10 -0.18
    INFO: ngram_search.c(466): Resized backpointer table to 10000 entries
    INFO: ngram_search.c(474): Resized score stack to 200000 entries
    INFO: ngram_search.c(466): Resized backpointer table to 20000 entries
    INFO: ngram_search.c(474): Resized score stack to 400000 entries
    INFO: ngram_search.c(466): Resized backpointer table to 40000 entries
    INFO: ngram_search.c(474): Resized score stack to 800000 entries
    INFO: ngram_search_fwdtree.c(1549): 30327 words recognized (3/fr)
    INFO: ngram_search_fwdtree.c(1551): 5937782 senones evaluated (660/fr)
    INFO: ngram_search_fwdtree.c(1553): 12516655 channels searched (1390/fr),
    809105 1st, 501258 last
    INFO: ngram_search_fwdtree.c(1557): 38953 words for which last channels
    evaluated (4/fr)
    INFO: ngram_search_fwdtree.c(1560): 194042 candidate words for entering last
    phone (21/fr)
    INFO: ngram_search_fwdtree.c(1562): fwdtree 14.33 CPU 0.159 xRT
    INFO: ngram_search_fwdtree.c(1565): fwdtree 14.34 wall 0.159 xRT
    INFO: ngram_search_fwdflat.c(305): Utterance vocabulary contains 169 words
    INFO: ngram_search_fwdflat.c(940): 9455 words recognized (1/fr)
    INFO: ngram_search_fwdflat.c(942): 1102467 senones evaluated (123/fr)
    INFO: ngram_search_fwdflat.c(944): 667149 channels searched (74/fr)
    INFO: ngram_search_fwdflat.c(946): 52360 words searched (5/fr)
    INFO: ngram_search_fwdflat.c(948): 15434 word transitions (1/fr)
    INFO: ngram_search_fwdflat.c(951): fwdflat 2.08 CPU 0.023 xRT
    INFO: ngram_search_fwdflat.c(954): fwdflat 2.08 wall 0.023 xRT
    INFO: ngram_search.c(1201): not found in last frame, using Agtyl.8997
    instead
    INFO: ngram_search.c(1253): lattice start node .0 end node Agtyl.2007
    INFO: ngram_search.c(1281): Eliminated 810 nodes before end node
    INFO: ngram_search.c(1386): Lattice has 1155 nodes, 423 links
    INFO: ps_lattice.c(1352): Normalizer P(O) = alpha(Agtyl:2007:8997) = -6836004
    INFO: ps_lattice.c(1390): Joint P(O,S) = -6836871 P(S|O) = -867
    INFO: ngram_search.c(875): bestpath 0.00 CPU 0.000 xRT
    INFO: ngram_search.c(878): bestpath 0.00 wall 0.000 xRT
    INFO: ps_lattice.c(242): Writing lattice file: /home/mehmet/pocket-
    sphinx/lattices/tmp_00m_00s__01m_30s.ready.lat
    *** glibc detected *** /home/mehmet/pocketsphinx/bin/pocketsphinx_batch: double free or corruption (out): 0x0000000003f10ed0 ***
    ======= Backtrace: =========
    /lib/libc.so.6(+0x775b6)
    /lib/libc.so.6(cfree+0x73)
    /home/mehmet/pocketsphinx/lib/libpocketsphinx.so.1(ps_astar_finish+0x20)
    /home/mehmet/pocketsphinx/bin/pocketsphinx_batch
    /home/mehmet/pocketsphinx/bin/pocketsphinx_batch
    /home/mehmet/pocketsphinx/bin/pocketsphinx_batch
    /lib/libc.so.6(__libc_start_main+0xfd)
    /home/mehmet/pocketsphinx/bin/pocketsphinx_batch
    ======= Memory map: ========
    00400000-00408000 r-xp 00000000 08:05 11636091
    /home/mehmet/pocketsphinx/bin/pocketsphinx_batch
    00607000-00608000 r--p 00007000 08:05 11636091
    /home/mehmet/pocketsphinx/bin/pocketsphinx_batch
    00608000-00609000 rw-p 00008000 08:05 11636091
    /home/mehmet/pocketsphinx/bin/pocketsphinx_batch
    0090f000-07d22000 rw-p 00000000 00:00 0
    7efef4000000-7efef4021000 rw-p 00000000 00:00 0
    7efef4021000-7efef8000000 ---p 00000000 00:00 0
    7efefa532000-7efefa548000 r-xp 00000000 08:01 6291535 /lib/libgcc_s.so.1
    7efefa548000-7efefa747000 ---p 00016000 08:01 6291535 /lib/libgcc_s.so.1
    7efefa747000-7efefa748000 r--p 00015000 08:01 6291535 /lib/libgcc_s.so.1
    7efefa748000-7efefa749000 rw-p 00016000 08:01 6291535 /lib/libgcc_s.so.1
    7efefa749000-7efefb0a1000 r--s 00000000 08:05 11668805 /home/mehmet/sphinx/lm
    /ar-ipsos/arabic.ug.lm.DMP
    7efefb0a1000-7efefb123000 r-xp 00000000 08:01 6291984 /lib/libm-2.11.1.so
    7efefb123000-7efefb322000 ---p 00082000 08:01 6291984 /lib/libm-2.11.1.so
    7efefb322000-7efefb323000 r--p 00081000 08:01 6291984 /lib/libm-2.11.1.so
    7efefb323000-7efefb324000 rw-p 00082000 08:01 6291984 /lib/libm-2.11.1.so
    7efefb324000-7efefb33c000 r-xp 00000000 08:01 6292004
    /lib/libpthread-2.11.1.so
    7efefb33c000-7efefb53b000 ---p 00018000 08:01 6292004
    /lib/libpthread-2.11.1.so
    7efefb53b000-7efefb53c000 r--p 00017000 08:01 6292004
    /lib/libpthread-2.11.1.so
    7efefb53c000-7efefb53d000 rw-p 00018000 08:01 6292004
    /lib/libpthread-2.11.1.so
    7efefb53d000-7efefb541000 rw-p 00000000 00:00 0
    7efefb541000-7efefb6bb000 r-xp 00000000 08:01 6291603 /lib/libc-2.11.1.so
    7efefb6bb000-7efefb8ba000 ---p 0017a000 08:01 6291603 /lib/libc-2.11.1.so
    7efefb8ba000-7efefb8be000 r--p 00179000 08:01 6291603 /lib/libc-2.11.1.so
    7efefb8be000-7efefb8bf000 rw-p 0017d000 08:01 6291603 /lib/libc-2.11.1.so
    7efefb8bf000-7efefb8c4000 rw-p 00000000 00:00 0
    7efefb8c4000-7efefb90c000 r-xp 00000000 08:05 11635519
    /home/mehmet/sphinxbase/lib/libsphinxbase.so.1.1.1
    7efefb90c000-7efefbb0c000 ---p 00048000 08:05 11635519
    /home/mehmet/sphinxbase/lib/libsphinxbase.so.1.1.1
    7efefbb0c000-7efefbb0d000 r--p 00048000 08:05 11635519
    /home/mehmet/sphinxbase/lib/libsphinxbase.so.1.1.1
    7efefbb0d000-7efefbb0f000 rw-p 00049000 08:05 11635519
    /home/mehmet/sphinxbase/lib/libsphinxbase.so.1.1.1
    7efefbb0f000-7efefbb10000 rw-p 00000000 00:00 0
    7efefbb10000-7efefbb4f000 r-xp 00000000 08:05 11636087
    /home/mehmet/pocketsphinx/lib/libpocketsphinx.so.1.1.0
    7efefbb4f000-7efefbd4f000 ---p 0003f000 08:05 11636087
    /home/mehmet/pocketsphinx/lib/libpocketsphinx.so.1.1.0
    7efefbd4f000-7efefbd51000 r--p 0003f000 08:05 11636087
    /home/mehmet/pocketsphinx/lib/libpocketsphinx.so.1.1.0
    7efefbd51000-7efefbd52000 rw-p 00041000 08:05 11636087
    /home/mehmet/pocketsphinx/lib/libpocketsphinx.so.1.1.0
    7efefbd52000-7efefbd72000 r-xp 00000000 08:01 6291500 /lib/ld-2.11.1.so
    7efefbf14000-7efefbf48000 rw-p 00000000 00:00 0
    7efefbf6c000-7efefbf71000 rw-p 00000000 00:00 0
    7efefbf71000-7efefbf72000 r--p 0001f000 08:01 6291500 /lib/ld-2.11.1.so
    7efefbf72000-7efefbf73000 rw-p 00020000 08:01 6291500 /lib/ld-2.11.1.so
    7efefbf73000-7efefbf74000 rw-p 00000000 00:00 0
    7fff4a453000-7fff4a468000 rw-p 00000000 00:00 0
    7fff4a5b3000-7fff4a5b4000 r-xp 00000000 00:00 0
    ffffffffff600000-ffffffffff601000 r-xp 00000000 00:00 0
    Aborted

    How can I get rid of this error and decode all the segments?

    Thanks in advance.

     
  • Nickolay V. Shmyrev

    What version are you using?

    How exactly do you run your application?

    Where can we get your data files to reproduce your problems?

    Please provide more details

     
  • mehmet

    mehmet - 2011-09-13

    I am using sphinxbase0.7 and pocketsphinx0.7, I guess they are the latest
    versions.
    I run pocketsphinx_batch with the following command:
    /home/mehmet/pocketsphinx/bin/pocketsphinx_batch -hmm /home/mehmet/sphinx/am
    /ar-ipsos -dict /home/mehmet/sphinx/lm/ar-ipsos/arabic.dic -lm
    /home/mehmet/sphinx/lm/ar-ipsos/arabic.ug.lm.DMP -cepdir /home/mehmet/pocket-
    sphinx/cep -ctl /home/mehmet/pocket-sphinx/files.dat -cepext .wav -adcin yes
    -samprate 16000 -hyp /home/mehmet/pocket-sphinx/out.hyp -nbestdir /home/mehmet
    /pocket-sphinx/nbest -nbest 10 -outlatdir /home/mehmet/pocket-sphinx/lattices
    Model files took more than 250MB, so I may upload them as a last resort.

     
  • Nickolay V. Shmyrev

    Sorry, It's hard to reproduce this problem and thus hard to fix it. I can only
    suggest you to download and try pocketsphinx snapshot to make sure that was
    not an old bug

    Time to time, audio files to be decoded may be too large to be decoded by
    PocketSphix, and I had to split them by using mp3splt toolkit.

    You need to use continuous speech recognition mode.

     

Log in to post a comment.