Hi!
I tried to use HTS modeling HSMMS for recognition. Because HTK doesn't support HSMMs.
I did HCompv, HInit, HRest well.
But, I got a problem using HERest.
I followed the training demo script.
demo script:
$HERest{'mon'} = "$HEREST -A -C $cfg{'trn'} -D -T 1 -S $scp{'trn'} -I $mlf{'mon'} -m 1 -u tmvwdmv -w $wf -t $beam ";
shell("$HERest{'mon'} -H $monommf{'cmp'} -N $monommf{'dur'} -M $model{'cmp'} -R $model{'dur'} $lst{'mon'} $lst{'mon'}");
I did like below:
/home/wkseong/HTK/HTS_2_1_1/bin/HERest -C ./config_tr_MFCC_E_D_A -D -T 1 -S ./TEST_file/Feature_DataBase/GIST_student_DB/Adapt_GIST_student_39.scp -I ./TEST_file/Lables/Keyword_monophone.mlf -m 1 -u tmvwdmv -w 3 -t 250.0 150.0 1000.0 -H ./HMM/dir2/hmm$j/model -N ./HMM/dir2/dur$j/dur -M ./HMM/dir2/hmm$i -R ./HMM/dir2/dur$i ./TEST_file/Lib/monophones ./TEST_file/Lib/monophones
but, any updating didn't happen, no error occured. When i did it for only HMM without duration model, it worked well.
what's the problem? Please help me!