HInit -A -D -C ../config -S de.hcompv.scp -M hmm0 -H hmm_hello -l hallo -L de/lab/ hallo

HTK Configuration Parameters[16]
  Module/Tool     Parameter                  Value
#                 NUMCEPS                       12
#                 CEPLIFTER                     22
#                 PREEMCOEF               0.970000
#                 USEHAMMING                  TRUE
#                 WINDOWSIZE         250000.000000
#                 SAVEWITHCRC                 TRUE
#                 SAVECOMPRESSED              TRUE
#                 TARGETRATE                100000
#                 TARGETFORMAT                 HTK
#                 TARGETKIND                MFCC_0
#                 USER                      MFCC_0
#                 ESCALE                  1.000000
#                 NUMCHANS                      26
#                 HEADERSIZE                     0
#                 SOURCERATE                   625
#                 SOURCEFORMAT            WAVEFORM

  ERROR [+2121]  HInit: Too Few Observation Sequences [0]
 FATAL ERROR - Terminating program HInit

now, i get this:

HInit -A -D -C hinit.conf -S de.hcompv.scp -M hmm0 -H proto -l hallo -L de/wav/ hallo

  ERROR [+2121]  HInit: Too Few Observation Sequences [1]
 FATAL ERROR - Terminating program HInit

you can follow my progress at: http://dbase.center/data/speech/

----

~o <VecSize> 13 <MFCC_0>
~h "hallo"
<BeginHMM>
<NumStates> 5
<State> 2
<Mean> 13
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
<Variance> 13
1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0
<State> 3
<Mean> 13
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
<Variance> 13
1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0
<State> 4
<Mean> 13
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
<Variance> 13
1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0
<TransP> 5
0.0 1.0 0.0 0.0 0.0
0.0 0.6 0.4 0.0 0.0
0.0 0.0 0.6 0.4 0.0
0.0 0.0 0.0 0.7 0.3
0.0 0.0 0.0 0.0 0.0
<EndHMM>

You do not have enough data for training, you need to add more training data. Ideally you need at least 30 samples for each word.