VoxForge
VoxForge
Hello there,
Still, I cannot deal with my .wav files generated by Audacity in HTK. My HTK can only deal with the .sig files generated by HSlab.
Since I need to do comparison experiment, I recorded my voice through audacity followed the recording toturials.
I have my sample data and all the additional files when running HCopy and HCompV.
I really need help:"( If I cannot solve the problem, I can't do my comparison experiment:"(
Who can help me? Can I send all my sample files to anyone else? Is my HTK wrong??
Thank you all for your great help!:"(
--- (Edited on 5/1/2012 6:45 pm [GMT-0500] by NancyZxll) ---
--- (Edited on 5/1/2012 6:46 pm [GMT-0500] by NancyZxll) ---
>I cannot deal with my .wav files generated by Audacity in HTK. My HTK
>can only deal with the .sig files generated by HSlab.
which O/S are you using? which version of HTK? were there any errors when you compiled your HTK?
--- (Edited on 5/1/2012 9:13 pm [GMT-0400] by kmaclean) ---
Hello Kmaclean,
I am using Windows, 32bits.
The HTK is the "current stable release" version: 3.4.1
I compiled my HTK according to the tutorial: http://htk.eng.cam.ac.uk/docs/inst-win.shtml
There was no error happend when compiling, but when I test my installation, I run
cd HTKDemo
perl runDemo.pl configs/monPlainM1S1.dcf
I didn't see the result shown in the tutorial. My result is very long, the last several paras were:
------------------------------------------------------------------
- system is PLAIN
90 Observation Sequences Loaded
Starting Estimation Process
Iteration 1: Average LogP = -531.87140
Iteration 2: Average LogP = -527.31677 Change = 4.55465
Iteration 3: Average LogP = -526.60223 Change = 0.71452
Iteration 4: Average LogP = -526.32343 Change = 0.27880
Iteration 5: Average LogP = -526.26605 Change = 0.05737
Iteration 6: Average LogP = -526.24451 Change = 0.02152
Iteration 7: Average LogP = -526.23560 Change = 0.00892
Iteration 8: Average LogP = -526.23358 Change = 0.00200
Iteration 9: Average LogP = -526.22858 Change = 0.00502
Iteration 10: Average LogP = -526.22864 Change = -0.00007
Estimation converged at iteration 11
Output written to directory hmms/hmm.0
HTK Configuration Parameters[3]
Module/Tool Parameter Value
KEEPDISTINCT FALSE
SAVEGLOBOPTS TRUE
TARGETKIND MFCC_E_D
Calling HInit for HMM V
HInit -A -i 10 -L labels\bcplabs\mon -l V -o V -C toolconfs\hinit.conf -D
s\hmm.0 -T 1 -S train.scr proto/V
HTK Configuration Parameters[3]
Module/Tool Parameter Value
# KEEPDISTINCT FALSE
# SAVEGLOBOPTS TRUE
# TARGETKIND MFCC_E_D
Initialising HMM proto/V . . .
States : 2 3 4 (width)
Mixes s1: 1 1 1 ( 26 )
Num Using: 0 0 0
Parm Kind: MFCC_E_D
Number of owners = 1
SegLab : V
maxIter : 10
epsilon : 0.000100
minSeg : 3
Updating : Means Variances MixWeights/DProbs TransProbs
- system is PLAIN
68 Observation Sequences Loaded
Starting Estimation Process
Iteration 1: Average LogP = -599.68665
Iteration 2: Average LogP = -592.98315 Change = 6.70348
Iteration 3: Average LogP = -592.21436 Change = 0.76877
Iteration 4: Average LogP = -591.48163 Change = 0.73274
Iteration 5: Average LogP = -591.14313 Change = 0.33848
Iteration 6: Average LogP = -591.08783 Change = 0.05529
Iteration 7: Average LogP = -591.04993 Change = 0.03791
Iteration 8: Average LogP = -590.98993 Change = 0.05998
Iteration 9: Average LogP = -590.98560 Change = 0.00435
Iteration 10: Average LogP = -590.98560 Change = 0.00001
Estimation converged at iteration 11
Output written to directory hmms/hmm.0
HTK Configuration Parameters[3]
Module/Tool Parameter Value
KEEPDISTINCT FALSE
SAVEGLOBOPTS TRUE
TARGETKIND MFCC_E_D
Calling HInit for HMM N
HInit -A -i 10 -L labels\bcplabs\mon -l N -o N -C toolconfs\hinit.conf -D
s\hmm.0 -T 1 -S train.scr proto/N
HTK Configuration Parameters[3]
Module/Tool Parameter Value
# KEEPDISTINCT FALSE
# SAVEGLOBOPTS TRUE
# TARGETKIND MFCC_E_D
Initialising HMM proto/N . . .
States : 2 3 4 (width)
Mixes s1: 1 1 1 ( 26 )
Num Using: 0 0 0
Parm Kind: MFCC_E_D
Number of owners = 1
SegLab : N
maxIter : 10
epsilon : 0.000100
minSeg : 3
Updating : Means Variances MixWeights/DProbs TransProbs
- system is PLAIN
16 Observation Sequences Loaded
Starting Estimation Process
Iteration 1: Average LogP = -398.18735
Iteration 2: Average LogP = -395.48715 Change = 2.70020
Iteration 3: Average LogP = -395.26947 Change = 0.21768
Iteration 4: Average LogP = -395.17661 Change = 0.09286
Iteration 5: Average LogP = -395.17661 Change = 0.00000
Estimation converged at iteration 6
Output written to directory hmms/hmm.0
HTK Configuration Parameters[3]
Module/Tool Parameter Value
KEEPDISTINCT FALSE
SAVEGLOBOPTS TRUE
TARGETKIND MFCC_E_D
Calling HInit for HMM L
HInit -A -i 10 -L labels\bcplabs\mon -l L -o L -C toolconfs\hinit.conf -D
s\hmm.0 -T 1 -S train.scr proto/L
HTK Configuration Parameters[3]
Module/Tool Parameter Value
# KEEPDISTINCT FALSE
# SAVEGLOBOPTS TRUE
# TARGETKIND MFCC_E_D
Initialising HMM proto/L . . .
States : 2 3 4 (width)
Mixes s1: 1 1 1 ( 26 )
Num Using: 0 0 0
Parm Kind: MFCC_E_D
Number of owners = 1
SegLab : L
maxIter : 10
epsilon : 0.000100
minSeg : 3
Updating : Means Variances MixWeights/DProbs TransProbs
- system is PLAIN
25 Observation Sequences Loaded
Starting Estimation Process
Iteration 1: Average LogP = -424.39636
Iteration 2: Average LogP = -420.31281 Change = 4.08355
Iteration 3: Average LogP = -419.43472 Change = 0.87808
Iteration 4: Average LogP = -419.35437 Change = 0.08035
Iteration 5: Average LogP = -419.32385 Change = 0.03050
Iteration 6: Average LogP = -419.32385 Change = -0.00001
Estimation converged at iteration 7
Output written to directory hmms/hmm.0
HTK Configuration Parameters[3]
Module/Tool Parameter Value
KEEPDISTINCT FALSE
SAVEGLOBOPTS TRUE
TARGETKIND MFCC_E_D
Can't open hmms\hmm.1
--- (Edited on 5/2/2012 11:54 am [GMT-0500] by NancyZxll) ---
>I didn't see the result shown in the tutorial.
Have you seen these posts: Compiling HTK 3.4 on Windows XP
have you tried installing HTK under cygwin?
--- (Edited on 5/2/2012 1:38 pm [GMT-0400] by kmaclean) ---
Hello Kmaclean,
So, you mean the HTK can deal with .wav files under Linux, right?
I have installed Linux VMare Machine on my windows. If the HTK under Linux windows can do it successfully, I will try to install the HTK again on my Linux VMare Machine.
Thank you so much Kmaclean!
--- (Edited on 5/2/2012 1:48 pm [GMT-0500] by Visitor) ---