Menu

SphinxTrain-1.0 running error

Help
blueskych
2010-06-30
2012-09-22
  • blueskych

    blueskych - 2010-06-30

    when I run "RunAll.pl", the result is:

    E:\VoiceTrain\SphinxTrain-1.0\SphinxTrain-1.0\scripts_pl>perl RunAll.pl
    MODULE: 00 verify training files
    O.S. is case insensitive ("A" == "a").
    Phones will be treated as case insensitive.
    Phase 1: DICT - Checking to see if the dict and filler dict agrees with the
    phonelist file.
    WARNING: Phone "N " has extra white spaces
    WARNING: Phone "OW " has extra white spaces
    WARNING: Phone "R " has extra white spaces
    WARNING: Phone "S " has extra white spaces
    WARNING: Phone "UW " has extra white spaces
    WARNING: Phone "V " has extra white spaces
    WARNING: The phonelist
    (E:\VoiceTrain\SphinxTrain-1.0\SphinxTrain-1.0/etc/time.p
    hone) does not define the phone SIL (required!)
    Found 11 words using 17 phones
    WARNING: This phone (N) occurs in the dictionary
    (E:\VoiceTrain\SphinxTrain-1.0\
    SphinxTrain-1.0/etc/time.dic), but not in the phonelist
    (E:\VoiceTrain\SphinxTra
    in-1.0\SphinxTrain-1.0/etc/time.phone)
    WARNING: This phone (OW) occurs in the dictionary
    (E:\VoiceTrain\SphinxTrain-1.0
    \SphinxTrain-1.0/etc/time.dic), but not in the phonelist
    (E:\VoiceTrain\SphinxTr
    ain-1.0\SphinxTrain-1.0/etc/time.phone)
    WARNING: This phone (R) occurs in the dictionary
    (E:\VoiceTrain\SphinxTrain-1.0\
    SphinxTrain-1.0/etc/time.dic), but not in the phonelist
    (E:\VoiceTrain\SphinxTra
    in-1.0\SphinxTrain-1.0/etc/time.phone)
    WARNING: This phone (S) occurs in the dictionary
    (E:\VoiceTrain\SphinxTrain-1.0\
    SphinxTrain-1.0/etc/time.dic), but not in the phonelist
    (E:\VoiceTrain\SphinxTra
    in-1.0\SphinxTrain-1.0/etc/time.phone)
    WARNING: This phone (SIL) occurs in the dictionary
    (E:\VoiceTrain\SphinxTrain-1.
    0\SphinxTrain-1.0/etc/time.dic), but not in the phonelist
    (E:\VoiceTrain\SphinxT
    rain-1.0\SphinxTrain-1.0/etc/time.phone)
    WARNING: This phone (UW) occurs in the dictionary
    (E:\VoiceTrain\SphinxTrain-1.0
    \SphinxTrain-1.0/etc/time.dic), but not in the phonelist
    (E:\VoiceTrain\SphinxTr
    ain-1.0\SphinxTrain-1.0/etc/time.phone)
    WARNING: This phone (V) occurs in the dictionary
    (E:\VoiceTrain\SphinxTrain-1.0\
    SphinxTrain-1.0/etc/time.dic), but not in the phonelist
    (E:\VoiceTrain\SphinxTra
    in-1.0\SphinxTrain-1.0/etc/time.phone)
    WARNING: This phone (N ) occurs in the phonelist
    (E:\VoiceTrain\SphinxTrain-1.0\
    SphinxTrain-1.0/etc/time.phone), but not in the dictionary
    (E:\VoiceTrain\Sphinx
    Train-1.0\SphinxTrain-1.0/etc/time.dic)
    WARNING: This phone (OW ) occurs in the phonelist
    (E:\VoiceTrain\SphinxTrain-1.0
    \SphinxTrain-1.0/etc/time.phone), but not in the dictionary
    (E:\VoiceTrain\Sphin
    xTrain-1.0\SphinxTrain-1.0/etc/time.dic)
    WARNING: This phone (R ) occurs in the phonelist
    (E:\VoiceTrain\SphinxTrain-1.0\
    SphinxTrain-1.0/etc/time.phone), but not in the dictionary
    (E:\VoiceTrain\Sphinx
    Train-1.0\SphinxTrain-1.0/etc/time.dic)
    WARNING: This phone (S ) occurs in the phonelist
    (E:\VoiceTrain\SphinxTrain-1.0\
    SphinxTrain-1.0/etc/time.phone), but not in the dictionary
    (E:\VoiceTrain\Sphinx
    Train-1.0\SphinxTrain-1.0/etc/time.dic)
    WARNING: This phone (UW ) occurs in the phonelist
    (E:\VoiceTrain\SphinxTrain-1.0
    \SphinxTrain-1.0/etc/time.phone), but not in the dictionary
    (E:\VoiceTrain\Sphin
    xTrain-1.0\SphinxTrain-1.0/etc/time.dic)
    WARNING: This phone (V ) occurs in the phonelist
    (E:\VoiceTrain\SphinxTrain-1.0\
    SphinxTrain-1.0/etc/time.phone), but not in the dictionary
    (E:\VoiceTrain\Sphinx
    Train-1.0\SphinxTrain-1.0/etc/time.dic)
    Phase 2: DICT - Checking to make sure there are not duplicate entries in the
    dictionary
    Phase 3: CTL - Check general format; utterance length (must be positive); fi
    les exist
    Phase 4: CTL - Checking number of lines in the transcript should match lines
    in control file
    Phase 5: CTL - Determine amount of training data, see if n_tied_states seems
    reasonable.
    Total Hours Training: 0.00119594017094017
    This is a small amount of data, no comment at this time
    Phase 6: TRANSCRIPT - Checking that all the words in the transcript are in t
    he dictionary
    Words in dictionary: 7
    Words in filler dictionary: 4
    Phase 7: TRANSCRIPT - Checking that all the phones in the transcript are in
    the phonelist, and all phones in the phonelist appear at least once
    WARNING: This phone (AH) occurs in the phonelist
    (E:\VoiceTrain\SphinxTrain-1.0\
    SphinxTrain-1.0/etc/time.phone), but not in any word in the transcription
    (E:\Vo
    iceTrain\SphinxTrain-1.0\SphinxTrain-1.0/etc/time.transcription)
    WARNING: This phone (AY) occurs in the phonelist
    (E:\VoiceTrain\SphinxTrain-1.0\
    SphinxTrain-1.0/etc/time.phone), but not in any word in the transcription
    (E:\Vo
    iceTrain\SphinxTrain-1.0\SphinxTrain-1.0/etc/time.transcription)
    WARNING: This phone (F) occurs in the phonelist
    (E:\VoiceTrain\SphinxTrain-1.0\S
    phinxTrain-1.0/etc/time.phone), but not in any word in the transcription
    (E:\Voi
    ceTrain\SphinxTrain-1.0\SphinxTrain-1.0/etc/time.transcription)
    WARNING: This phone (IH) occurs in the phonelist
    (E:\VoiceTrain\SphinxTrain-1.0\
    SphinxTrain-1.0/etc/time.phone), but not in any word in the transcription
    (E:\Vo
    iceTrain\SphinxTrain-1.0\SphinxTrain-1.0/etc/time.transcription)
    WARNING: This phone (IY) occurs in the phonelist
    (E:\VoiceTrain\SphinxTrain-1.0\
    SphinxTrain-1.0/etc/time.phone), but not in any word in the transcription
    (E:\Vo
    iceTrain\SphinxTrain-1.0\SphinxTrain-1.0/etc/time.transcription)
    WARNING: This phone (K) occurs in the phonelist
    (E:\VoiceTrain\SphinxTrain-1.0\S
    phinxTrain-1.0/etc/time.phone), but not in any word in the transcription
    (E:\Voi
    ceTrain\SphinxTrain-1.0\SphinxTrain-1.0/etc/time.transcription)
    WARNING: This phone (N ) occurs in the phonelist
    (E:\VoiceTrain\SphinxTrain-1.0\
    SphinxTrain-1.0/etc/time.phone), but not in any word in the transcription
    (E:\Vo
    iceTrain\SphinxTrain-1.0\SphinxTrain-1.0/etc/time.transcription)
    WARNING: This phone (OW ) occurs in the phonelist
    (E:\VoiceTrain\SphinxTrain-1.0
    \SphinxTrain-1.0/etc/time.phone), but not in any word in the transcription
    (E:\V
    oiceTrain\SphinxTrain-1.0\SphinxTrain-1.0/etc/time.transcription)
    WARNING: This phone (R ) occurs in the phonelist
    (E:\VoiceTrain\SphinxTrain-1.0\
    SphinxTrain-1.0/etc/time.phone), but not in any word in the transcription
    (E:\Vo
    iceTrain\SphinxTrain-1.0\SphinxTrain-1.0/etc/time.transcription)
    WARNING: This phone (S ) occurs in the phonelist
    (E:\VoiceTrain\SphinxTrain-1.0\
    SphinxTrain-1.0/etc/time.phone), but not in any word in the transcription
    (E:\Vo
    iceTrain\SphinxTrain-1.0\SphinxTrain-1.0/etc/time.transcription)
    WARNING: This phone (T) occurs in the phonelist
    (E:\VoiceTrain\SphinxTrain-1.0\S
    phinxTrain-1.0/etc/time.phone), but not in any word in the transcription
    (E:\Voi
    ceTrain\SphinxTrain-1.0\SphinxTrain-1.0/etc/time.transcription)
    WARNING: This phone (TH) occurs in the phonelist
    (E:\VoiceTrain\SphinxTrain-1.0\
    SphinxTrain-1.0/etc/time.phone), but not in any word in the transcription
    (E:\Vo
    iceTrain\SphinxTrain-1.0\SphinxTrain-1.0/etc/time.transcription)
    WARNING: This phone (UW ) occurs in the phonelist
    (E:\VoiceTrain\SphinxTrain-1.0
    \SphinxTrain-1.0/etc/time.phone), but not in any word in the transcription
    (E:\V
    oiceTrain\SphinxTrain-1.0\SphinxTrain-1.0/etc/time.transcription)
    WARNING: This phone (V ) occurs in the phonelist
    (E:\VoiceTrain\SphinxTrain-1.0\
    SphinxTrain-1.0/etc/time.phone), but not in any word in the transcription
    (E:\Vo
    iceTrain\SphinxTrain-1.0\SphinxTrain-1.0/etc/time.transcription)
    WARNING: This phone (W) occurs in the phonelist
    (E:\VoiceTrain\SphinxTrain-1.0\S
    phinxTrain-1.0/etc/time.phone), but not in any word in the transcription
    (E:\Voi
    ceTrain\SphinxTrain-1.0\SphinxTrain-1.0/etc/time.transcription)
    WARNING: This phone (Z) occurs in the phonelist
    (E:\VoiceTrain\SphinxTrain-1.0\S
    phinxTrain-1.0/etc/time.phone), but not in any word in the transcription
    (E:\Voi
    ceTrain\SphinxTrain-1.0\SphinxTrain-1.0/etc/time.transcription)
    Something failed:
    (E:\VoiceTrain\SphinxTrain-1.0\SphinxTrain-1.0/scripts_pl/00.v
    erify/verify_all.pl)

    and my .dic file is:
    0 Z IY R OW
    1 W AH N
    2 T UW
    3 TH R IY
    4 F OW R
    5 F AY V
    6 S IH K S
    my
    .transcription file is:
    0
    1
    2
    3
    4
    5
    6
    my *.phone file is:
    AH
    AY
    F
    IH
    IY
    K
    N
    OW
    R
    S
    T
    TH
    UW
    V
    W
    Z

    I don't konw what is wrong?

     
  • Nickolay V. Shmyrev

    I don't konw what is wrong?

    This message in the log you pasted

    WARNING: Phone "N " has extra white spaces

    tells you that your phone file has extra spaces. In your phone file those
    spaces are obviously not visible, but nevertheless they should be deleted
    before you'll start the training.

    I have also comment about your transcription file. Obviosly, 10 recordings is
    not enough to train the model, you will not have luck with such a small amount
    of data. I suggest you to use more recordings for training.

     
  • Nickolay V. Shmyrev

    And your transcription format is not correct btw. Transcription should have
    utterance id in braces on each line

    text (file_name_without_extension)
    another text (another_file_name_without_extension)

    I suggest you to look at tutorial database first.

     
  • Alain SIMAC-LEJEUNE

    It's same for me
    when I run "RunAll.pl", the result is:

    root@al1:/home/litii/SPHINX# ./scripts_pl/RunAll.pl
    MODULE: 00 verify training files
    O.S. is case sensitive ("A" != "a").
    Phones will be treated as case sensitive.
    Phase 1: DICT - Checking to see if the dict and filler dict agrees with the
    phonelist file.
    " has extra white spaces
    " has extra white spaces
    " has extra white spaces
    " has extra white spaces
    " has extra white spaces
    " has extra white spaces
    " has extra white spaces
    " has extra white spaces
    " has extra white spaces
    " has extra white spaces
    " has extra white spaces
    " has extra white spaces
    " has extra white spaces
    " has extra white spaces
    " has extra white spaces
    " has extra white spaces
    " has extra white spaces
    " has extra white spaces
    " has extra white spaces
    WARNING: The phonelist (/home/litii/SPHINX/etc/chiffres.phone) does not define
    the phone SIL (required!)
    Found 16 words using 19 phones
    WARNING: This phone (SIL) occurs in the dictionary
    (/home/litii/SPHINX/etc/chiffres.dic), but not in the phonelist
    (/home/litii/SPHINX/etc/chiffres.phone)
    WARNING: This phone (a) occurs in the dictionary
    (/home/litii/SPHINX/etc/chiffres.dic), but not in the phonelist
    (/home/litii/SPHINX/etc/chiffres.phone)
    WARNING: This phone (ai) occurs in the dictionary
    (/home/litii/SPHINX/etc/chiffres.dic), but not in the phonelist
    (/home/litii/SPHINX/etc/chiffres.phone)
    WARNING: This phone (d) occurs in the dictionary
    (/home/litii/SPHINX/etc/chiffres.dic), but not in the phonelist
    (/home/litii/SPHINX/etc/chiffres.phone)
    WARNING: This phone (e) occurs in the dictionary
    (/home/litii/SPHINX/etc/chiffres.dic), but not in the phonelist
    (/home/litii/SPHINX/etc/chiffres.phone)
    WARNING: This phone (eu) occurs in the dictionary
    (/home/litii/SPHINX/etc/chiffres.dic), but not in the phonelist
    (/home/litii/SPHINX/etc/chiffres.phone)
    WARNING: This phone (f) occurs in the dictionary
    (/home/litii/SPHINX/etc/chiffres.dic), but not in the phonelist
    (/home/litii/SPHINX/etc/chiffres.phone)
    WARNING: This phone (i) occurs in the dictionary
    (/home/litii/SPHINX/etc/chiffres.dic), but not in the phonelist
    (/home/litii/SPHINX/etc/chiffres.phone)
    WARNING: This phone (in) occurs in the dictionary
    (/home/litii/SPHINX/etc/chiffres.dic), but not in the phonelist
    (/home/litii/SPHINX/etc/chiffres.phone)
    WARNING: This phone (k) occurs in the dictionary
    (/home/litii/SPHINX/etc/chiffres.dic), but not in the phonelist
    (/home/litii/SPHINX/etc/chiffres.phone)
    WARNING: This phone (n) occurs in the dictionary
    (/home/litii/SPHINX/etc/chiffres.dic), but not in the phonelist
    (/home/litii/SPHINX/etc/chiffres.phone)
    WARNING: This phone (o) occurs in the dictionary
    (/home/litii/SPHINX/etc/chiffres.dic), but not in the phonelist
    (/home/litii/SPHINX/etc/chiffres.phone)
    WARNING: This phone (r) occurs in the dictionary
    (/home/litii/SPHINX/etc/chiffres.dic), but not in the phonelist
    (/home/litii/SPHINX/etc/chiffres.phone)
    WARNING: This phone (s) occurs in the dictionary
    (/home/litii/SPHINX/etc/chiffres.dic), but not in the phonelist
    (/home/litii/SPHINX/etc/chiffres.phone)
    WARNING: This phone (t) occurs in the dictionary
    (/home/litii/SPHINX/etc/chiffres.dic), but not in the phonelist
    (/home/litii/SPHINX/etc/chiffres.phone)
    WARNING: This phone (un) occurs in the dictionary
    (/home/litii/SPHINX/etc/chiffres.dic), but not in the phonelist
    (/home/litii/SPHINX/etc/chiffres.phone)
    WARNING: This phone (w) occurs in the dictionary
    (/home/litii/SPHINX/etc/chiffres.dic), but not in the phonelist
    (/home/litii/SPHINX/etc/chiffres.phone)
    WARNING: This phone (y) occurs in the dictionary
    (/home/litii/SPHINX/etc/chiffres.dic), but not in the phonelist
    (/home/litii/SPHINX/etc/chiffres.phone)
    WARNING: This phone (z) occurs in the dictionary
    (/home/litii/SPHINX/etc/chiffres.dic), but not in the phonelist
    (/home/litii/SPHINX/etc/chiffres.phone)
    ) occurs in the phonelist (/home/SPHINX/etc/chiffres.phone), but not in the
    dictionary (/home/SPHINX/etc/chiffres.dic)
    ) occurs in the phonelist (/home/SPHINX/etc/chiffres.phone), but not in the
    dictionary (/home/SPHINX/etc/chiffres.dic)
    ) occurs in the phonelist (/home/SPHINX/etc/chiffres.phone), but not in the
    dictionary (/home/SPHINX/etc/chiffres.dic)
    ) occurs in the phonelist (/home/SPHINX/etc/chiffres.phone), but not in the
    dictionary (/home/SPHINX/etc/chiffres.dic)
    ) occurs in the phonelist (/home/SPHINX/etc/chiffres.phone), but not in the
    dictionary (/home/SPHINX/etc/chiffres.dic)
    ) occurs in the phonelist (/home/SPHINX/etc/chiffres.phone), but not in the
    dictionary (/home/SPHINX/etc/chiffres.dic)
    ) occurs in the phonelist (/home/SPHINX/etc/chiffres.phone), but not in the
    dictionary (/home/SPHINX/etc/chiffres.dic)
    ) occurs in the phonelist (/home/SPHINX/etc/chiffres.phone), but not in the
    dictionary (/home/SPHINX/etc/chiffres.dic)
    ) occurs in the phonelist (/home/SPHINX/etc/chiffres.phone), but not in the
    dictionary (/home/SPHINX/etc/chiffres.dic)
    ) occurs in the phonelist (/home/SPHINX/etc/chiffres.phone), but not in the
    dictionary (/home/SPHINX/etc/chiffres.dic)
    ) occurs in the phonelist (/home/SPHINX/etc/chiffres.phone), but not in the
    dictionary (/home/SPHINX/etc/chiffres.dic)
    ) occurs in the phonelist (/home/SPHINX/etc/chiffres.phone), but not in the
    dictionary (/home/SPHINX/etc/chiffres.dic)
    ) occurs in the phonelist (/home/SPHINX/etc/chiffres.phone), but not in the
    dictionary (/home/SPHINX/etc/chiffres.dic)
    ) occurs in the phonelist (/home/SPHINX/etc/chiffres.phone), but not in the
    dictionary (/home/SPHINX/etc/chiffres.dic)
    ) occurs in the phonelist (/home/SPHINX/etc/chiffres.phone), but not in the
    dictionary (/home/SPHINX/etc/chiffres.dic)
    ) occurs in the phonelist (/home/SPHINX/etc/chiffres.phone), but not in the
    dictionary (/home/SPHINX/etc/chiffres.dic)
    ) occurs in the phonelist (/home/SPHINX/etc/chiffres.phone), but not in the
    dictionary (/home/SPHINX/etc/chiffres.dic)
    ) occurs in the phonelist (/home/SPHINX/etc/chiffres.phone), but not in the
    dictionary (/home/SPHINX/etc/chiffres.dic)
    ) occurs in the phonelist (/home/SPHINX/etc/chiffres.phone), but not in the
    dictionary (/home/SPHINX/etc/chiffres.dic)
    Phase 2: DICT - Checking to make sure there are not duplicate entries in the
    dictionary
    Phase 3: CTL - Check general format; utterance length (must be positive);
    files exist
    Phase 4: CTL - Checking number of lines in the transcript should match lines
    in control file
    Phase 5: CTL - Determine amount of training data, see if n_tied_states seems
    reasonable.
    Total Hours Training: 0.0976465811965812
    This is a small amount of data, no comment at this time
    Phase 6: TRANSCRIPT - Checking that all the words in the transcript are in the
    dictionary
    Words in dictionary: 13
    Words in filler dictionary: 3
    Phase 7: TRANSCRIPT - Checking that all the phones in the transcript are in
    the phonelist, and all phones in the phonelist appear at least once
    ) occurs in the phonelist (/home/SPHINX/etc/chiffres.phone), but not in any
    word in the transcription (/home/SPHINX/etc/chiffres_train.transcription)
    ) occurs in the phonelist (/home/SPHINX/etc/chiffres.phone), but not in any
    word in the transcription (/home/SPHINX/etc/chiffres_train.transcription)
    ) occurs in the phonelist (/home/SPHINX/etc/chiffres.phone), but not in any
    word in the transcription (/home/SPHINX/etc/chiffres_train.transcription)
    ) occurs in the phonelist (/home/SPHINX/etc/chiffres.phone), but not in any
    word in the transcription (/home/SPHINX/etc/chiffres_train.transcription)
    ) occurs in the phonelist (/home/SPHINX/etc/chiffres.phone), but not in any
    word in the transcription (/home/SPHINX/etc/chiffres_train.transcription)
    ) occurs in the phonelist (/home/SPHINX/etc/chiffres.phone), but not in any
    word in the transcription (/home/SPHINX/etc/chiffres_train.transcription)
    ) occurs in the phonelist (/home/SPHINX/etc/chiffres.phone), but not in any
    word in the transcription (/home/SPHINX/etc/chiffres_train.transcription)
    ) occurs in the phonelist (/home/SPHINX/etc/chiffres.phone), but not in any
    word in the transcription (/home/SPHINX/etc/chiffres_train.transcription)
    ) occurs in the phonelist (/home/SPHINX/etc/chiffres.phone), but not in any
    word in the transcription (/home/SPHINX/etc/chiffres_train.transcription)
    ) occurs in the phonelist (/home/SPHINX/etc/chiffres.phone), but not in any
    word in the transcription (/home/SPHINX/etc/chiffres_train.transcription)
    ) occurs in the phonelist (/home/SPHINX/etc/chiffres.phone), but not in any
    word in the transcription (/home/SPHINX/etc/chiffres_train.transcription)
    ) occurs in the phonelist (/home/SPHINX/etc/chiffres.phone), but not in any
    word in the transcription (/home/SPHINX/etc/chiffres_train.transcription)
    ) occurs in the phonelist (/home/SPHINX/etc/chiffres.phone), but not in any
    word in the transcription (/home/SPHINX/etc/chiffres_train.transcription)
    ) occurs in the phonelist (/home/SPHINX/etc/chiffres.phone), but not in any
    word in the transcription (/home/SPHINX/etc/chiffres_train.transcription)
    ) occurs in the phonelist (/home/SPHINX/etc/chiffres.phone), but not in any
    word in the transcription (/home/SPHINX/etc/chiffres_train.transcription)
    ) occurs in the phonelist (/home/SPHINX/etc/chiffres.phone), but not in any
    word in the transcription (/home/SPHINX/etc/chiffres_train.transcription)
    ) occurs in the phonelist (/home/SPHINX/etc/chiffres.phone), but not in any
    word in the transcription (/home/SPHINX/etc/chiffres_train.transcription)
    ) occurs in the phonelist (/home/SPHINX/etc/chiffres.phone), but not in any
    word in the transcription (/home/SPHINX/etc/chiffres_train.transcription)
    ) occurs in the phonelist (/home/SPHINX/etc/chiffres.phone), but not in any
    word in the transcription (/home/SPHINX/etc/chiffres_train.transcription)
    Something failed: (/home/SPHINX/scripts_pl/00.verify/verify_all.pl)

    chiffres.phone
    a
    ai
    d
    e
    eu
    f
    i
    in
    k
    n
    o
    r
    s
    t
    un
    w
    y
    z

    chiffres.dic
    cinq s in
    cinq(2) s in k
    deux d eu
    huit y i
    huit(2) y i t
    neuf n eu f
    quatre k a t r
    sept s e t
    six s i
    six(2) s i s
    trois t r w a
    un un
    zero z ai r o

    chiffres.transcription
    un deux trois quatre un (audio1)
    sept huit neuf neuf deux (audio2)
    quatre cinq six cinq sept (audio3)
    un quatre sept cinq six (audio4)
    six neuf deux quatre huit (audio5)
    huit quatre cinq six deux (audio6)
    deux trois neuf zero cinq (audio7)
    zero un neuf huit quatre (audio8)
    sept zero sept huit neuf (audio9)
    neuf un trois deux zero (audio10)
    trois sept trois zero deux (audio11)
    huit six sept zero un (audio12)
    deux trois six un zero (audio13)
    zero un deux un zero (audio14)
    trois quatre cinq quatre trois (audio15)
    six sept huit sept six (audio16)
    huit neuf zero sept huit (audio17)
    cinq quatre six quatre deux (audio18)
    cinq neuf zero neuf cinq (audio19)
    quatre trois deux un zero (audio20)
    neuf huit sept six cinq (audio21)
    six cinq quatre trois deux (audio22)
    trois huit sept sept neuf (audio23)
    neuf trois quatre deux un (audio24)
    un huit zero deux huit (audio25)
    un un trois cinq six (audio26)
    zero (audio27)
    un (audio28)
    deux (audio29)
    trois (audio30)
    quatre (audio31)
    cinq (audio32)
    six (audio33)
    sept (audio34)
    huit (audio35)
    neuf (audio36)

    I don't konw what is wrong?

     
  • Nickolay V. Shmyrev

    I don't konw what is wrong?

    You didn't read the message training script provided you with the description
    of the problem. It said your phoneset file has extra whitespaces.

     
  • Alain SIMAC-LEJEUNE

    I have modified the file for eliminated the white space.
    But :

    root@litii-al1:/home/ASR/chiffres# ./scripts_pl/RunAll.pl
    MODULE: 00 verify training files
    O.S. is case sensitive ("A" != "a").
    Phones will be treated as case sensitive.
    Phase 1: DICT - Checking to see if the dict and filler dict agrees with the
    phonelist file.
    Found 16 words using 19 phones
    Phase 2: DICT - Checking to make sure there are not duplicate entries in the
    dictionary
    Phase 3: CTL - Check general format; utterance length (must be positive);
    files exist
    Phase 4: CTL - Checking number of lines in the transcript should match lines
    in control file
    Phase 5: CTL - Determine amount of training data, see if n_tied_states seems
    reasonable.
    Total Hours Training: 0.0976465811965812
    This is a small amount of data, no comment at this time
    Phase 6: TRANSCRIPT - Checking that all the words in the transcript are in the
    dictionary
    Words in dictionary: 13
    Words in filler dictionary: 3
    Phase 7: TRANSCRIPT - Checking that all the phones in the transcript are in
    the phonelist, and all phones in the phonelist appear at least once
    WARNING: This phone (A) occurs in the phonelist
    (/home/ASR/chiffres/etc/chiffres.phone), but not in any word in the
    transcription (/home/ASR/chiffres/etc/chiffres_train.transcription)
    WARNING: This phone (AI) occurs in the phonelist
    (/home/ASR/chiffres/etc/chiffres.phone), but not in any word in the
    transcription (/home/ASR/chiffres/etc/chiffres_train.transcription)
    WARNING: This phone (D) occurs in the phonelist
    (/home/ASR/chiffres/etc/chiffres.phone), but not in any word in the
    transcription (/home/ASR/chiffres/etc/chiffres_train.transcription)
    WARNING: This phone (E) occurs in the phonelist
    (/home/ASR/chiffres/etc/chiffres.phone), but not in any word in the
    transcription (/home/ASR/chiffres/etc/chiffres_train.transcription)
    WARNING: This phone (EU) occurs in the phonelist
    (/home/ASR/chiffres/etc/chiffres.phone), but not in any word in the
    transcription (/home/ASR/chiffres/etc/chiffres_train.transcription)
    WARNING: This phone (F) occurs in the phonelist
    (/home/ASR/chiffres/etc/chiffres.phone), but not in any word in the
    transcription (/home/ASR/chiffres/etc/chiffres_train.transcription)
    WARNING: This phone (I) occurs in the phonelist
    (/home/ASR/chiffres/etc/chiffres.phone), but not in any word in the
    transcription (/home/ASR/chiffres/etc/chiffres_train.transcription)
    WARNING: This phone (IN) occurs in the phonelist
    (/home/ASR/chiffres/etc/chiffres.phone), but not in any word in the
    transcription (/home/ASR/chiffres/etc/chiffres_train.transcription)
    WARNING: This phone (K) occurs in the phonelist
    (/home/litii/ASR/chiffres/etc/chiffres.phone), but not in any word in the
    transcription (/home/ASR/chiffres/etc/chiffres_train.transcription)
    WARNING: This phone (N) occurs in the phonelist
    (/home/ASR/chiffres/etc/chiffres.phone), but not in any word in the
    transcription (/home/ASR/chiffres/etc/chiffres_train.transcription)
    WARNING: This phone (O) occurs in the phonelist
    (/home/ASR/chiffres/etc/chiffres.phone), but not in any word in the
    transcription (/home/ASR/chiffres/etc/chiffres_train.transcription)
    WARNING: This phone (R) occurs in the phonelist
    (/home/ASR/chiffres/etc/chiffres.phone), but not in any word in the
    transcription (/home/ASR/chiffres/etc/chiffres_train.transcription)
    WARNING: This phone (S) occurs in the phonelist
    (/home/ASR/chiffres/etc/chiffres.phone), but not in any word in the
    transcription (/home/ASR/chiffres/etc/chiffres_train.transcription)
    WARNING: This phone (SIL) occurs in the phonelist
    (/home/ASR/chiffres/etc/chiffres.phone), but not in any word in the
    transcription (/home/ASR/chiffres/etc/chiffres_train.transcription)
    WARNING: This phone (T) occurs in the phonelist
    (/home/ASR/chiffres/etc/chiffres.phone), but not in any word in the
    transcription (/home/ASR/chiffres/etc/chiffres_train.transcription)
    WARNING: This phone (UN) occurs in the phonelist
    (/home/ASR/chiffres/etc/chiffres.phone), but not in any word in the
    transcription (/home/ASR/chiffres/etc/chiffres_train.transcription)
    WARNING: This phone (W) occurs in the phonelist
    (/home/ASR/chiffres/etc/chiffres.phone), but not in any word in the
    transcription (/home/ASR/chiffres/etc/chiffres_train.transcription)
    WARNING: This phone (Y) occurs in the phonelist
    (/home/ASR/chiffres/etc/chiffres.phone), but not in any word in the
    transcription (/home/ASR/chiffres/etc/chiffres_train.transcription)
    WARNING: This phone (Z) occurs in the phonelist
    (/home/ASR/chiffres/etc/chiffres.phone), but not in any word in the
    transcription (/home/ASR/chiffres/etc/chiffres_train.transcription)
    MODULE: 01 Vector Quantization

    And error in module 20 Phase 2 - Flat initialize because no data

    The exemple an4 of tutorial compiled.
    My work not.
    The file .phone and .transcription is same to an4.

    .phone :
    A
    AI
    D
    E
    EU
    F
    I
    IN
    K
    N
    O
    R
    S
    SIL
    T
    UN
    W
    Y
    Z

    .transcription :
    UN DEUX TROIS QUATRE UN A (audio1)

    SEPT HUIT NEUF NEUF DEUX (audio2)

    QUATRE CINQ SIX CINQ SEPT (audio3)

    UN QUATRE SEPT CINQ SIX (audio4)

    SIX NEUF DEUX QUATRE HUIT (audio5)

    HUIT QUATRE CINQ SIX DEUX (audio6)

    DEUX TROIS NEUF ZERO CINQ (audio7)

    ZERO UN NEUF HUIT QUATRE (audio8)

    SEPT ZERO SEPT HUIT NEUF (audio9)

    NEUF UN TROIS DEUX ZERO (audio10)

    TROIS SEPT TROIS ZERO DEUX (audio11)

    HUIT SIX SEPT ZERO UN (audio12)

    DEUX TROIS SIX UN ZERO (audio13)

    ZERO UN DEUX UN ZERO (audio14)

    TROIS QUATRE CUBQ QUATRE TROIS (audio15)

    SIX SEPR HUIT SEPT SIX (audio16)

    HUIT NEUF ZERO SEPT HUIT (audio17)

    CINQ QUATRE SIX QUATRE DEUX (audio18)

    CINQ NUEF ZERO NEUF CINQ (audio19)

    QUATRE TROIS DEUX UN ZERO (audio20)

    NEUF HUIT SEPT SIX CINQ (audio21)

    SIX CINQ QUATRE TROIS DEUX (audio22)

    TROIS HUIT SEPT SEPT NEUF (audio23)

    NEUF TROIS QUATRE DEUX UN (audio24)

    UN HUIT ZERO DEUX HUIT (audio25)

    UN UN TROIS CINQ SIX (audio26)

    ZERO (audio27)

    UN (audio28)

    DEUX (audio29)

    TROIS (audio30)

    QUATRE (audio31)

    CINQ (audio32)

    SIX (audio33)

    SEPT (audio34

     
  • Nickolay V. Shmyrev

    Your dictionary is lower case. Your phoneset is upper case. It wrote you

    Phones will be treated as case sensitive.
    This phone (K) occurs in the phonelist
    (/home/litii/ASR/chiffres/etc/chiffres.phone), but not in any word in the
    transcription (/home/ASR/chiffres/etc/chiffres_train.transcription

     
  • Alain SIMAC-LEJEUNE

    For the first message, my dictionnary is lower case but now, all file is UPPER
    CASE (cf message 2)

    WARNING: This phone (Z) occurs in the phonelist
    (/home/ASR/chiffres/etc/chiffres.phone), but not in any word in the
    transcription (/home/ASR/chiffres/etc/chiffres_train.transcription)
    .phone :
    Z
    ZERO UN DEUX UN ZERO (audio14)

     
  • Nickolay V. Shmyrev

    Another possible reason is that your transcription file has extra spaces in
    the end of the line. Try to cleanup them

    In case it doesn't work try to follow the FAQ suggestion. You need to share
    your configuration files. You need to pack them into archive and upload to a
    public sharing resource. Don't forget to give a link on uploaded archive after
    that.

     
  • Alain SIMAC-LEJEUNE

    I have verify the files (tab, spaces, encoding format) but nothing different
    to an4 example.

    My file : http://www.simaclejeune.fr/chiffres.zip

    Thanks for the help

     
  • Nickolay V. Shmyrev

    Your transcription does have extra whitespaces in the end so as your fileids
    file. They are windows-style newlines CR. You need to clean them with dos2unix
    for example.

    Please read the tutorial carefully before you start training.

    http://cmusphinx.sourceforge.net/wiki/tutorialam

     
  • Steve

    Steve - 2011-04-06

    I am getting the exact same warning, which I think is causing more issues
    further down the line. Have you (alainsl) or anybody else figured out what
    causes this?
    Thanks.

     
  • Steve

    Steve - 2011-04-06

    Please ignore previous post. For some reason none of the replys were showing
    up.

     
  • sudip

    sudip - 2011-10-31

    I've already installed festival 1.9.5.

    According to the tinydoc of the Sphinx Train("If you decide to create your own
    dictionary, create it with (only
    works for US English)

    bin/make_dict etc/time.transcription

    This will create etc/word.known etc/word.unknown files; check them").

    BUT I can`t find bin/make_dict in festival ,

    I don't know why.

    Now I need your help.thank you very much! looking forward your reply.
    how can i make my own dictionary

     
  • Nickolay V. Shmyrev

    The script location is sphinxtrain/scripts_pl/make_dict.festival. It's a part
    of the SphinxTrain package. You need to use festival --script to run it.

     
  • sudip

    sudip - 2011-11-03

    ODULE: 05 Build Trees (2011-11-02 11:04)
    Cleaning up old log files...
    Make Questions
    Log File

    WARN: "e:\tan 10et90r03\sphinxtrain\src\libs\libio\s3io.c", line 247: Unable
    to open c:/SphinxTrain/time/model_parameters/time.ci_cont/mixture_weights for
    reading; FATAL_ERROR: "d:\sphinxtrain\src\programs\make_quests\main.c", line
    1060: Initialization failed
    FAILED
    MODULE: 06 Prune Trees (2011-11-02 11:30)
    Cleaning up old log files...
    mk_mdef_gen Log File completed
    Prune trees
    Log File

    FATAL: "d:\sphinxtrain\src\programs\prunetree\main.c", line 167: Unable to
    open c:/SphinxTrain/time/trees/time.unpruned/AH-0.dtree for reading;
    FAILED
    what are the solutions for this two error..though i have a very small
    database..one ,one(2)and two

     
  • Nickolay V. Shmyrev

    what are the solutions for this two error

    Read and understand the tutorial

    Use latest SphinxTrain

    If you have an issue, provide the whole build folder, not just a little line
    from the logs.

    Read the logs before you submit them

    Understand what is going on

     

Log in to post a comment.