VoxForge
VoxForge
I have prepared all model needed to run Julius. The contents of jconf file (julius.jconf) to run julius are:
-nlr model/grammar # 2-gram
-v dict
-h hmm15/hmmdefs
-hlist tiedlist
-gprune safe # safe pruning �ãˆÊNŒÂ‚ªŠm�À‚É‹�‚Ü‚é�D�³Šm�D
-n 10
-output 10 # ‘æ2ƒpƒX‚ÅŒ©‚‚©‚Á‚½•¶‚Ì‚¤‚¿�o—Í‚·‚é�” �i•¶�”�j
-input rawfile # ‰¹�º”gŒ`ƒf�[ƒ^ƒtƒ@ƒCƒ‹(ƒtƒH�[ƒ}ƒbƒg�©“®”»•Ê)
# Œ`�®�FWAV(16bit) ‚Ü‚½‚Í
# RAW(16bit(signed short),mono,big-endian)
# 16kHzˆÈŠO‚̃tƒ@ƒCƒ‹‚Í -smpFreq ‚Å�ü”g�”�w’è
-filelist listfile.txt # ”F�¯‘Î�Ûƒtƒ@ƒCƒ‹‚̃ŠƒXƒg
-zmean # DC�¬•ª‚Ì�œ‹�‚ð�s‚¤ (-input mfcfile��–³Œø)
-rejectshort 100 # �w’èƒ~ƒŠ•bˆÈ‰º‚Ì’·‚³‚Ì“ü—Í‚ðŠü‹p‚·‚é
-lv 10000 # ƒŒƒxƒ‹‚Ì‚µ‚«‚¢’l (0-32767)
-smpFreq 16000 # ƒTƒ“ƒvƒŠƒ“ƒO�ü”g�”(Hz)
-smpPeriod 625 # ƒTƒ“ƒvƒŠƒ“ƒO�üŠú(ns) (= 10000000 / smpFreq)
-demo # "-progout -quiet" ‚Æ“¯‚¶
When I run julius by typing:
$julius -C julius.jconf
There were errors as follows:
STAT: include config: julius.jconf
STAT: jconf successfully finalized
STAT: *** loading AM00 _default
Stat: init_phmm: Reading in HMM definition
Stat: rdhmmdef: ascii format HMM definition
Stat: rdhmmdef: limit check passed
Stat: check_hmm_restriction: an HMM with several arcs from initial state found:
"sp"
Stat: rdhmmdef: this HMM requires multipath handling at decoding
Stat: init_phmm: defined HMMs: 17
Stat: init_phmm: logical names: 46 in HMMList
Stat: init_phmm: base phones: 17 used in logical
Stat: init_phmm: finished reading HMM definitions
STAT: making pseudo bi/mono-phone for IW-triphone
Stat: hmm_lookup: 28 pseudo phones are added to logical HMM list
STAT: *** AM00 _default loaded
STAT: *** loading LM00 _default
Error: voca_load_htkdict: line 1: triphone "uh-ah+sp" not found
Error: voca_load_htkdict: line 1: triphone "ah-sp+*" or biphone "ah-sp" not foun
d
Error: voca_load_htkdict: the line content was: DUA [DUA]
d uh ah sp
Error: voca_load_htkdict: line 2: triphone "ah-t+sp" not found
Error: voca_load_htkdict: line 2: triphone "t-sp+*" or biphone "t-sp" not found
Error: voca_load_htkdict: the line content was: EMPAT [EMPAT]
ax m p ah t sp
Error: voca_load_htkdict: line 3: triphone "ah-m+sp" not found
Error: voca_load_htkdict: line 3: triphone "m-sp+*" or biphone "m-sp" not found
Error: voca_load_htkdict: the line content was: ENAM [ENAM]
ax n ah m sp
Error: voca_load_htkdict: line 4: triphone "oh-ng+sp" not found
Error: voca_load_htkdict: line 4: triphone "ng-sp+*" or biphone "ng-sp" not foun
d
Error: voca_load_htkdict: the line content was: KOSONG [KOSONG]
k oh sh oh ng sp
Error: voca_load_htkdict: line 5: triphone "m-ah+sp" not found
Error: voca_load_htkdict: line 5: triphone "ah-sp+*" or biphone "ah-sp" not foun
d
Error: voca_load_htkdict: the line content was: LIMA [LIMA]
l ih m ah sp
Error: voca_load_htkdict: line 6: triphone "t-uh+sp" not found
Error: voca_load_htkdict: line 6: triphone "uh-sp+*" or biphone "uh-sp" not foun
d
Error: voca_load_htkdict: the line content was: SATU [SATU]
sh ah t uh sp
Error: voca_load_htkdict: line 9: triphone "g-ah+sp" not found
Error: voca_load_htkdict: line 9: triphone "ah-sp+*" or biphone "ah-sp" not foun
d
Error: voca_load_htkdict: the line content was: TIGA [TIGA]
t ih g ah sp
Error: voca_load_htkdict: begin missing phones
Error: voca_load_htkdict: ah-m+sp
Error: voca_load_htkdict: ah-sp+* or biphone ah-sp
Error: voca_load_htkdict: ah-t+sp
Error: voca_load_htkdict: g-ah+sp
Error: voca_load_htkdict: m-ah+sp
Error: voca_load_htkdict: m-sp+* or biphone m-sp
Error: voca_load_htkdict: ng-sp+* or biphone ng-sp
Error: voca_load_htkdict: oh-ng+sp
Error: voca_load_htkdict: t-sp+* or biphone t-sp
Error: voca_load_htkdict: t-uh+sp
Error: voca_load_htkdict: uh-ah+sp
Error: voca_load_htkdict: uh-sp+* or biphone uh-sp
Error: voca_load_htkdict: end missing phones
Error: init_voca: error in reading dict: 7 words failed out of 2 words
ERROR: m_fusion: failed to read dictionary, terminated
ERROR: m_fusion: failed to initialize dictionary
ERROR: Error in loading model
I knew that those missing phones didn't exist in "triphones1" file, but this file was generated by HTKToolkit. I think this problem occurs because of the content of "dict" file that consists of list of words like this: "DUA [DUA] d uh ah sp", which is generated by HTKToolkit and already consists of "sp". Can you help me to solve this problem? Thank you very much.
regards,
Amalia Zahra
My first question above is about the problem when I executed julius by using 2 gram language model (grammar). But, when I executed julian by using sample.dfa and sample.dict, the cygwin displayed the result as follows:
Stat: adin_file: input speechfile: wav_indonesia/1406-answer-00.wav
Warning: strip: sample 33944-33960 is invalid, stripped
Warning: strip: sample 42532-42548 is invalid, stripped
Warning: strip: sample 60120-60136 is invalid, stripped
Warning: strip: sample 123304-123320 is invalid, stripped
Warning: strip: sample 132844-132860 is invalid, stripped
Warning: strip: sample 134184-134204 is invalid, stripped
STAT: 143571 samples (8.97 sec.)
STAT: ### speech analysis (waveform -> MFCC)
pass1_best: <s> LIMA EMPAT EMPAT DUA ENAM SATU TIGA TIGA TIGA TIGA
pass1_best: <s> LIMA EMPAT EMPAT DUA ENAM SATU TIGA TIGA TIGA TIGA LIMA ENAM SAT
U </s>
sentence1: <s> EMPAT SATU </s>
What is the meaning of warning above? How can I fix it? Then..when I ran julian with another wav input file, but the same content with wav file already trained, julius recognized it incorrectly. Does julius/julian only recognize wav input file that has been trained before? Can't julius/julian recognize other files never been trained, but consists of same content with wav files already trained before. I really need your help for my questions both. Thanks.
regards,
Amalia Zahra
Hi amza,
>My first question above is about the problem when I executed julius by using 2
>gram language model (grammar).
Sorry, I don't have much experience running Julius in dictation mode - I can't help you on this one.
>What is the meaning of warning above? How can I fix it?
Is there a problem with the audio in "1406-answer-00.wav"? Can you listen to the speech audio in an audio editor? Is there lots of background noise? Was it recorded with a good quality microphone?
Have you looked at the Julius/Julian code to see what might cause the error?
You might try emailing Julius support directly - I've never seen this error before.
>Does julius/julian only recognize wav input file that has been trained before?
No.
If you have created a good triphone acoustic model (using the steps in the Tutorial) Julian should recognize short phrases containing words that were not necessarily included in the training set.
However, when you don't have enough training data, then you might be limited to only the words you trained with.
Ken
Hi people,
I have the same problem than amza. (the first error)
Would you mind telling me how yo do to solve it, please?
Did you add the not found triphones at your tieldlist? If I did it, it wolud slove the problem?
I ask you because you had the same problem than me, but nobody answer your question, and some time later you had a different problem. I suppose that you solve the original error yourself.
Thanks.
Hi,
>I have the same problem than amza. (the first error)
You might find your answer by searching the entire HTK users or developers mailing lists. You can download the archives using these links (note: you need to be logged in to HTK to download these ...):
Ken
The problem is that you load the dictionary that the spelling contains sp (short pause). In the other hand, the model tied state triphone you use is not contain sp.
Try to use the dictionary that not contain sp.
I hope this helps you. Because your problem is same.
Hi,
I've a similar problem to Amalia one.
Some triphone are not recognized when I run julius after htk grammar building based on the voxforge tutorial.
Output is:
julius-4.1.2 -C ../Sample.jconf
STAT: include config: ../Sample.jconf
STAT: jconf successfully finalized
STAT: *** loading AM00 _default
Stat: init_phmm: Reading in HMM definition
Stat: rdhmmdef: ascii format HMM definition
Stat: rdhmmdef: limit check passed
Stat: check_hmm_restriction: an HMM with several arcs from initial state found: "sp"
Stat: rdhmmdef: this HMM requires multipath handling at decoding
Stat: init_phmm: defined HMMs: 542
Stat: init_phmm: loading ascii hmmlist
Stat: init_phmm: logical names: 2849 in HMMList
Stat: init_phmm: base phones: 35 used in logical
Stat: init_phmm: finished reading HMM definitions
STAT: making pseudo bi/mono-phone for IW-triphone
Stat: hmm_lookup: 492 pseudo phones are added to logical HMM list
STAT: *** AM00 _default loaded
STAT: *** loading LM00 _default
STAT: reading [/home/francesco/voxforge/manual/sample.dfa] and [/home/francesco/voxforge/manual/sample.dict]...
Error: voca_load_htkdict: line 3: triphone "k-j+a1" not found
Error: voca_load_htkdict: line 3: triphone "j-a1+m" not found
Error: voca_load_htkdict: line 3: triphone "a1-m+a" not found
Error: voca_load_htkdict: the line content was: 2 [chiama] k j a1 m a
Error: voca_load_htkdict: line 4: triphone "a-m+a1" not found
Error: voca_load_htkdict: line 4: triphone "m-a1+i" not found
Error: voca_load_htkdict: line 4: triphone "a1-i+*" or biphone "a1-i" not found
Error: voca_load_htkdict: the line content was: 2 [chiamai] k j a m a1 i
Error: voca_load_htkdict: line 5: triphone "a-m+a1" not found
Error: voca_load_htkdict: line 5: triphone "m-a1+n" not found
Error: voca_load_htkdict: line 5: triphone "a1-n+d" not found
Error: voca_load_htkdict: the line content was: 2 [chiamando] k j a m a1 n d o
Error: voca_load_htkdict: line 6: triphone "k-j+a1" not found
Error: voca_load_htkdict: line 6: triphone "j-a1+m" not found
Error: voca_load_htkdict: line 6: triphone "a1-m+a" not found
Error: voca_load_htkdict: the line content was: 2 [chiamano] k j a1 m a n o
Error: voca_load_htkdict: line 7: triphone "a-m+a1" not found
Error: voca_load_htkdict: line 7: triphone "m-a1+r" not found
Error: voca_load_htkdict: line 7: triphone "a1-r+e" not found
Error: voca_load_htkdict: the line content was: 2 [chiamare] k j a m a1 r e
Error: voca_load_htkdict: line 8: triphone "a-m+a1" not found
Error: voca_load_htkdict: line 8: triphone "m-a1+r" not found
Error: voca_load_htkdict: line 8: triphone "a1-r+o" not found
Error: voca_load_htkdict: the line content was: 2 [chiamarono] k j a m a1 r o n o
Error: voca_load_htkdict: line 9: triphone "a-m+a1" not found
Error: voca_load_htkdict: line 9: triphone "m-a1+s" not found
Error: voca_load_htkdict: line 9: triphone "a1-s+s" not found
Error: voca_load_htkdict: the line content was: 2 [chiamasse] k j a m a1 s s e
Error: voca_load_htkdict: line 10: triphone "a-m+a1" not found
Error: voca_load_htkdict: line 10: triphone "m-a1+t" not found
Error: voca_load_htkdict: line 10: triphone "a1-t+a" not found
Error: voca_load_htkdict: the line content was: 2 [chiamata] k j a m a1 t a
Error: voca_load_htkdict: line 11: triphone "a-m+a1" not found
Error: voca_load_htkdict: line 11: triphone "m-a1+t" not found
Error: voca_load_htkdict: line 11: triphone "a1-t+e" not found
Error: voca_load_htkdict: the line content was: 2 [chiamate] k j a m a1 t e
Error: voca_load_htkdict: line 12: triphone "a-m+a1" not found
Error: voca_load_htkdict: line 12: triphone "m-a1+t" not found
Error: voca_load_htkdict: line 12: triphone "a1-t+i" not found
Error: voca_load_htkdict: the line content was: 2 [chiamati] k j a m a1 t i
Error: voca_load_htkdict: line 13: triphone "a-m+a1" not found
Error: voca_load_htkdict: line 13: triphone "m-a1+t" not found
Error: voca_load_htkdict: line 13: triphone "a1-t+o" not found
Error: voca_load_htkdict: the line content was: 2 [chiamato] k j a m a1 t o
Error: voca_load_htkdict: line 14: triphone "a-m+a1" not found
Error: voca_load_htkdict: line 14: triphone "m-a1+v" not found
Error: voca_load_htkdict: line 14: triphone "a1-v+a" not found
Error: voca_load_htkdict: the line content was: 2 [chiamava] k j a m a1 v a
Error: voca_load_htkdict: line 15: triphone "a-m+a1" not found
Error: voca_load_htkdict: line 15: triphone "m-a1+v" not found
Error: voca_load_htkdict: line 15: triphone "a1-v+a" not found
Error: voca_load_htkdict: the line content was: 2 [chiamavano] k j a m a1 v a n o
Error: voca_load_htkdict: line 16: triphone "k-j+a1" not found
Error: voca_load_htkdict: line 16: triphone "j-a1+m" not found
Error: voca_load_htkdict: line 16: triphone "a1-m+i" not found
Error: voca_load_htkdict: the line content was: 2 [chiami] k j a1 m i
Error: voca_load_htkdict: line 17: triphone "a-m+j" not found
Error: voca_load_htkdict: line 17: triphone "m-j+a1" not found
Error: voca_load_htkdict: line 17: triphone "j-a1+m" not found
Error: voca_load_htkdict: line 17: triphone "a1-m+o" not found
Error: voca_load_htkdict: the line content was: 2 [chiamiamo] k j a m j a1 m o
Error: voca_load_htkdict: line 18: triphone "k-j+a1" not found
Error: voca_load_htkdict: line 18: triphone "j-a1+m" not found
Error: voca_load_htkdict: line 18: triphone "a1-m+o" not found
Error: voca_load_htkdict: the line content was: 2 [chiamo] k j a1 m o
Error: voca_load_htkdict: line 21: triphone "EE-f+o1" not found
Error: voca_load_htkdict: line 21: triphone "f-o1+n" not found
Error: voca_load_htkdict: line 21: triphone "o1-n+a" not found
Error: voca_load_htkdict: the line content was: 2 [telefona] t e l EE f o1 n a
Error: voca_load_htkdict: line 22: triphone "*-t+e1" or biphone "t+e1" not found
Error: voca_load_htkdict: line 22: triphone "t-e1+l" not found
Error: voca_load_htkdict: line 22: triphone "e1-l+e" not found
Error: voca_load_htkdict: line 22: triphone "e-f+o1" not found
Error: voca_load_htkdict: line 22: triphone "f-o1+n" not found
Error: voca_load_htkdict: line 22: triphone "o1-n+a" not found
Error: voca_load_htkdict: the line content was: 2 [telefona] t e1 l e f o1 n a
Error: voca_load_htkdict: line 24: triphone "m-p+o1" not found
Error: voca_load_htkdict: line 24: triphone "p-o1+n" not found
Error: voca_load_htkdict: line 24: triphone "o1-n+i" not found
Error: voca_load_htkdict: the line content was: 3 [componi] k o m p o1 n i
Error: voca_load_htkdict: line 25: triphone "m-p+o1" not found
Error: voca_load_htkdict: line 25: triphone "p-o1+r" not found
Error: voca_load_htkdict: line 25: triphone "o1-r+r" not found
Error: voca_load_htkdict: the line content was: 3 [comporre] k o m p o1 r r e
Error: voca_load_htkdict: line 26: triphone "e-l+i1" not found
Error: voca_load_htkdict: line 26: triphone "l-i1+s" not found
Error: voca_load_htkdict: line 26: triphone "i1-s+*" or biphone "i1-s" not found
Error: voca_load_htkdict: the line content was: 4 [angelis] a n dZZ e l i1 s
Error: voca_load_htkdict: line 27: triphone "*-a1+n" or biphone "a1+n" not found
Error: voca_load_htkdict: line 27: triphone "a1-n+dZZ" not found
Error: voca_load_htkdict: the line content was: 4 [angelis] a1 n dZZ e l i s
Error: voca_load_htkdict: line 28: triphone "r-s+a1" not found
Error: voca_load_htkdict: line 28: triphone "s-a1+r" not found
Error: voca_load_htkdict: line 28: triphone "a1-r+o" not found
Error: voca_load_htkdict: the line content was: 4 [corsaro] k OO r s a1 r o
Error: voca_load_htkdict: line 29: triphone "r-s+a1" not found
Error: voca_load_htkdict: line 29: triphone "s-a1+r" not found
Error: voca_load_htkdict: line 29: triphone "a1-r+o" not found
Error: voca_load_htkdict: the line content was: 4 [corsaro] k o r s a1 r o
Error: voca_load_htkdict: line 30: triphone "i-t+a1" not found
Error: voca_load_htkdict: line 30: triphone "t-a1+n" not found
Error: voca_load_htkdict: line 30: triphone "a1-n+o" not found
Error: voca_load_htkdict: the line content was: 4 [capitano] k a p i t a1 n o
Error: voca_load_htkdict: line 32: triphone "*-k+a1" or biphone "k+a1" not found
Error: voca_load_htkdict: line 32: triphone "k-a1+v" not found
Error: voca_load_htkdict: line 32: triphone "a1-v+a" not found
Error: voca_load_htkdict: the line content was: 4 [cavaliere] k a1 v a l j EE r e
Error: voca_load_htkdict: line 34: triphone "*-d+e1" or biphone "d+e1" not found
Error: voca_load_htkdict: line 34: triphone "d-e1+*" or biphone "d-e1" not found
Error: voca_load_htkdict: the line content was: 4 [de] d e1
Error: voca_load_htkdict: line 35: triphone "*-i1+l" or biphone "i1+l" not found
Error: voca_load_htkdict: line 35: triphone "i1-l+*" or biphone "i1-l" not found
Error: voca_load_htkdict: the line content was: 4 [il] i1 l
Error: voca_load_htkdict: line 38: triphone "*-d+u1" or biphone "d+u1" not found
Error: voca_load_htkdict: line 38: triphone "d-u1+e" not found
Error: voca_load_htkdict: line 38: triphone "u1-e+*" or biphone "u1-e" not found
Error: voca_load_htkdict: the line content was: 5 [due] d u1 e
Error: voca_load_htkdict: line 45: triphone "*-u1+n" or biphone "u1+n" not found
Error: voca_load_htkdict: line 45: triphone "u1-n+o" not found
Error: voca_load_htkdict: the line content was: 5 [uno] u1 n o
Error: voca_load_htkdict: begin missing phones
Error: voca_load_htkdict: *-a1+n or biphone a1+n
Error: voca_load_htkdict: *-d+e1 or biphone d+e1
Error: voca_load_htkdict: *-d+u1 or biphone d+u1
Error: voca_load_htkdict: *-i1+l or biphone i1+l
Error: voca_load_htkdict: *-k+a1 or biphone k+a1
Error: voca_load_htkdict: *-t+e1 or biphone t+e1
Error: voca_load_htkdict: *-u1+n or biphone u1+n
Error: voca_load_htkdict: EE-f+o1
Error: voca_load_htkdict: a-m+a1
Error: voca_load_htkdict: a-m+j
Error: voca_load_htkdict: a1-i+* or biphone a1-i
Error: voca_load_htkdict: a1-m+a
Error: voca_load_htkdict: a1-m+i
Error: voca_load_htkdict: a1-m+o
Error: voca_load_htkdict: a1-n+d
Error: voca_load_htkdict: a1-n+dZZ
Error: voca_load_htkdict: a1-n+o
Error: voca_load_htkdict: a1-r+e
Error: voca_load_htkdict: a1-r+o
Error: voca_load_htkdict: a1-s+s
Error: voca_load_htkdict: a1-t+a
Error: voca_load_htkdict: a1-t+e
Error: voca_load_htkdict: a1-t+i
Error: voca_load_htkdict: a1-t+o
Error: voca_load_htkdict: a1-v+a
Error: voca_load_htkdict: d-e1+* or biphone d-e1
Error: voca_load_htkdict: d-u1+e
Error: voca_load_htkdict: e-f+o1
Error: voca_load_htkdict: e-l+i1
Error: voca_load_htkdict: e1-l+e
Error: voca_load_htkdict: f-o1+n
Error: voca_load_htkdict: i-t+a1
Error: voca_load_htkdict: i1-l+* or biphone i1-l
Error: voca_load_htkdict: i1-s+* or biphone i1-s
Error: voca_load_htkdict: j-a1+m
Error: voca_load_htkdict: k-a1+v
Error: voca_load_htkdict: k-j+a1
Error: voca_load_htkdict: l-i1+s
Error: voca_load_htkdict: m-a1+i
Error: voca_load_htkdict: m-a1+n
Error: voca_load_htkdict: m-a1+r
Error: voca_load_htkdict: m-a1+s
Error: voca_load_htkdict: m-a1+t
Error: voca_load_htkdict: m-a1+v
Error: voca_load_htkdict: m-j+a1
Error: voca_load_htkdict: m-p+o1
Error: voca_load_htkdict: o1-n+a
Error: voca_load_htkdict: o1-n+i
Error: voca_load_htkdict: o1-r+r
Error: voca_load_htkdict: p-o1+n
Error: voca_load_htkdict: p-o1+r
Error: voca_load_htkdict: r-s+a1
Error: voca_load_htkdict: s-a1+r
Error: voca_load_htkdict: t-a1+n
Error: voca_load_htkdict: t-e1+l
Error: voca_load_htkdict: u1-e+* or biphone u1-e
Error: voca_load_htkdict: u1-n+o
Error: voca_load_htkdict: end missing phones
Error: init_voca: error in reading /home/francesco/voxforge/manual/sample.dict: 30 words failed out of 15 words
ERROR: failed to read dictionary "/home/francesco/voxforge/manual/sample.dict"
ERROR: m_fusion: some error occured in reading grammars
ERROR: Error in loading model
As you can see... it cannot fiind some triphone (I know that them are not present, but all files are created with htk like the tutorial).
Can you help me please??
Thanks,
littleJohn
adding information to my problem exposed in precedent post:
file dict (dictionary) used for building hmm net contain 'sp' (automatically added by htk tools); file sample.dic, dictionary used for grammar, NOT contains 'sp' pause..
Can anyone help me to solve this problem??
thank's in advance
best regards
Hi littleJohn,
>As you can see... it cannot fiind some triphone
You probably need more words in your training set - try to add 3 different words for each of the missing triphones.
Were there any errors in your step 10 commands?
It looks like you are trying to create a Spanish recognizer... are you using a Spanish tree.hed script (instead of the English on in Step 10)?
Ken
![[ ]](http://htk.eng.cam.ac.uk/icons/unknown.gif){kind=icon}