HTK-HCompv

HCompv

1.Function
  This program will calculate the global mean and covariance of a set of training data. It is primarily used to initialise the parameters of a HMM such that all component meand and all covatiances are set equal to the global data mean and covatiance. This might form the first start training scheme where all models are initially given the same parameters.

2. $HCompV -A -D -T 1 -C config -f 0.01 -m -S train.scp -M hmm0 proto

-C cf :set config file to cf
-f f:Output vFloor as f * global var
-c dir  Set output directiry for CMV         none
-f f    Output vFloor as f * global var      none
-k s    spkr pattern for CMV                 none
-l s    Set segment label to s               none
-m      Update means                         off
-o fn   Store new hmm def in fn (name only)  outDir/srcfn
-p s    path pattern for CMV                 none
-q nmv  output type flags for CMV            m
-v f    Set minimum variance to f            0.0
-A      Print command line arguments         off
-B      Save HMMs/transforms as binary       off
-C cf   Set config file to cf                default
-D      Display configuration variables      off
-F fmt  Set source data format to fmt        as config
-G fmt  Set source label format to fmt       as config
-H mmf  Load HMM macro file mmf
-I mlf  Load master label file mlf
-L dir  Set input label (or net) dir         current
-M dir  Dir to write HMM macro files         current
-S f    Set script file to f                 none
-T N    Set trace flags to N                 0
-V      Print version information            off
-X ext  Set input label (or net) file ext    lab