3.2.2 Step 7 - Fixing the Silence Models

 

The previous step has generated a 3 state left-to-right HMM for each phone and also a HMM for the silence model  sil. The next step is to add extra transitions from states 2 to 4 and from states 4 to 2  in the silence model. The idea here is to make the model more robust by allowing individual states to absorb the various impulsive noises in the training data. The backward skip allows this to happen without committing the model to transit to the following word.

Also, at this point, a 1 state short pause  sp model should be created. This should be a so-called tee-model  which has a direct transition from entry to exit node. This sp has its emitting state tied to the centre state of the silence model. The required topology of the two silence models is shown in Fig. 3.9.

These silence models can be created in two stages

• Use a text editor on the file hmm3/hmmdefs to copy the centre state of the sil model to make a new sp model and store the resulting MMF texttthmmdefs, which includes the new sp model, in the new directory hmm4.
• Run the HMM editor HHED  to add the extra transitions required and tie the sp state to the centre sil state

HHED works in a similar way to HLED. It applies a set of commands in a script to modify a set of HMMs. In this case, it is executed as follows

    HHEd -H hmm4/macros -H hmm4/hmmdefs -M hmm5 sil.hed monophones1

where sil.hed contains the following commands

    AT 2 4 0.2 {sil.transP}
    AT 4 2 0.2 {sil.transP}
    AT 1 3 0.3 {sp.transP}
    TI silst {sil.state[3],sp.state[2]}

The AT  commands add transitions to the given transition matrices and the final TI  command creates a tied-state called silst. The parameters of this tied-state are stored in the macros file and within each silence model, the original state parameters are replaced by the name of this macro . Macros are described in more detail below. For now it is sufficient to regard them simply as the mechanism by which HTK implements parameter sharing. Note that the phone list used here has been changed, because the original list monophones0 has been extended by the new sp model. The new file is called monophones1 and has been used in the above HHED command.

 

Finally, another two passes of HEREST are applied using the phone transcriptions with sp models between words. This leaves the set of monophone HMMs created so far in the directory hmm7. This step is illustrated in Fig. 3.10