From monophone hmms to tied-state triphones

this is down in two steps.
firstly, the monophone transcriptions are converted to triphone transcriptions and a set of triphones and a set of triphone models are created by copying the monophones and re-estimating,
secondly, similar acoustic states of these triphones are tied to ensure that all state distribution can be robustly estimated,

HLEd -n triphones1 -l '*' -i wintri.mlf mktri.led aligned.mlf
will convert the monophone transcription in aligned.mlf to an equivalent set of triphone transcription in wintri.mlf. At the same time, a list of triphones  is written to the file triphones1.
mktri.led contains the command
WB sp
WB sil
TC

HHEd -B -H hmm9/macros -H hmm9/hmmdefs -M hmm10 mktri.hed monophones1
mktri.hed
CL triphones1
TI T_ah{(*-ah+*, ah+*, *-ah).transP}
.....

the outcome if the previous stage is a set of triphone hmms with all triphones in a phone set sharing the same transition matrix. when estimating these models, many of the variances in the output distributions will have been floored since there will be insufficient data associated with many of the states. The last step in the model building process is to tie states within triphone sets in order to share data and thus be able to make robust parameter estimates.