Next: 13.10.4 Tracing
Up: 13.10 HLStats
Previous: 13.10.2 Bigram Generation
HLSTATS is invoked by the command line
HLStats [options] hmmList labFiles ....
The hmmList should contain a list of all the labels (ie model names)
used in the following label files for which statistics are required. Any
labels not appearing in the list are ignored and assumed to be
out-of-vocabulary. The list of labels is specified in the same way as for a
HMM list (see HMODEL) and the logical
physical mapping is preserved
to allow statistics to be collected about physical names as well as logical
ones. The labFiles may be master label files.
The available options are
- -b fn
- Compute bigram statistics and store result in the
file fn.
- -c N
- Count the number of occurrences of each logical model
listed in the hmmList and on completion list all models
for which there are N or less occurrences.
- -d
- Compute minimum, maximum and average duration statistics for each
label.
- -f f
- Set the matrix
bigram floor probability to f
(default value 0.0). This option should be used in
conjunction with the -b option.
- -h N
- Set the bigram hashtable size to medium(N=1) or
large (N=2). This option should be used in
conjunction with the -b option. The default is small(N=0).
- -l fn
- Output a list of covering labels to file fn.
Only labels occurring in the labList are counted (others
are assumed to be out-of-vocabulary).
However, this list may contain labels that do not occur in any of
the label files. The list of labels written to fn will however
contain only those labels which occur at least once.
- -o
- Produce backed-off bigrams rather than matrix ones. These
are output in the standard ARPA/MIT-LL textual format.
- -p N
- Count the number of occurrences of each physical model
listed in the hmmList and on completion list all models
for which there are N or less occurrences.
- -s st en
- Set the sentence start and end labels to st
and en. (Default !ENTER and !EXIT).
- -t n
- Set the threshold count for including a bigram
in a backed-off bigram language model. This option should be used in
conjunction with the -b and -o options.
- -u f
- Set the unigram floor probability to f when
constructing a back-off bigram language model.
This option should be used in
conjunction with the -b and -o options.
- -G fmt
- Set the label file format to fmt.
- -I mlf
- This loads the master label file mlf. This option
may be repeated to load several MLFs.
- -X ext
- Set label file extension to ext
(default is lab).
HLSTATS also supports the standard options -A,
-C, -D, -S, -T, and -V as described
in section 4.4.
Next: 13.10.4 Tracing
Up: 13.10 HLStats
Previous: 13.10.2 Bigram Generation
ECRL HTK_V2.1: email [email protected]