labels.tex

隐马尔科夫模型工具箱

the originals.  This is done by using the 
\texttt{-l} option.  For example,
\begin{verbatim}
    HLEd -l newlabs edfile.led l1 l2 l3
\end{verbatim}
would have the same effect as previously except that the new
label files would be stored in the directory \texttt{newlabs}.

Each edit command stored in an edit file\index{edit file} is identified by
a mnemonic consisting of two letters\footnote{
Some command names have single 
letter\index{edit commands!single letter} alternatives for compatibility with
earlier versions of \HTK.}
and must be stored on a separate
line.  The supplied edit commands can be divided into two groups.
The first group consist of commands which perform selective 
changes to specific labels and the second group contains commands
which perform global transformations.  
The reference section defines all of these
commands.  Here a few examples will be given to illustrate the
use of \htool{HLEd}.

As a first example, when using the 
TIMIT database\index{TIMIT database},  
the original 61 phoneme symbol set is often mapped
into a simpler 48 phoneme symbol set.  The aim of this mapping
is to delete all glottal stops, replace all closures preceding a voiced stop 
by a generic
voiced closure (\texttt{vcl}), all closures preceding an unvoiced stop 
by a generic
unvoiced closure (\texttt{cl}) and the different types of silence to
a single generic silence (\texttt{sil}).  A \htool{HLEd} script to
do this might be
\begin{verbatim}
    # Map 61 Phone Timit Set -> 48 Phones
    SO
    DE q
    RE cl pcl tcl kcl qcl
    RE vcl bcl dcl gcl
    RE sil h# #h pau
\end{verbatim}
The first line is a comment indicated by the initial hash character.
The command on the second line is the {\it Sort} command 
\texttt{SO}\index{so@\texttt{SO} command}.\index{labels!sorting}
This is an example of a global command.
Its effect is to sort all the labels into time order.
Normally the labels in a transcription will already be in time order
but some speech editors simply output labels in the order that the
transcriber marked them.  Since this would confuse the re-estimation
tools, it is good practice to explicitly sort all label files in this
way.

The command on the third line is the {\it Delete} command 
\texttt{DE}\index{de@\texttt{DE} command}.  
This is
a selective command.  Its\index{labels!deleting}
effect is to delete all of the labels
listed on the rest of the command line, wherever they occur.
In this case, there is
just one label listed for deletion, the
glottal stop  \texttt{q}.
Hence, the overall effect of this command will be to delete all occurrences
of the \texttt{q} label in the edited label files.

The remaining commands in this example script are {\it Replace} commands
\texttt{RE}.  The effect of a Replace command is to substitute the first
label following the \texttt{RE}\index{re@\texttt{RE} command} for every occurrence of the remaining labels
on that line.\index{labels!replacing}  Thus, for example, the 
command on the third line causes all occurrences of the labels \texttt{pcl},
\texttt{tcl}, \texttt{kcl} or \texttt{qcl} to be replaced by the label \texttt{cl}.

To illustrate the overall effect of the above \htool{HLEd} command script on
a complete label file, the following \texttt{TIMIT} format label file
\begin{verbatim}
     0000 2241 h#
     2241 2715 w
     2715 4360 ow
     4360 5478 bcl
     5478 5643 b
     5643 6360 iy
     6360 7269 tcl
     7269 8313 t
     8313 11400 ay
    11400 12950 dcl
    12950 14360 dh
    14360 14640 h#
\end{verbatim}
would be converted by the above script to the following
\begin{verbatim}
          0 1400625 sil 
    1400625 1696875 w 
    1696875 2725000 ow 
    2725000 3423750 vcl 
    3423750 3526875 b 
    3526875 3975000 iy 
    3975000 4543125 cl 
    4543125 5195625 t 
    5195625 7125000 ay 
    7125000 8093750 vcl 
    8093750 8975000 dh 
    8975000 9150000 sil 
\end{verbatim}
Notice that label boundaries in TIMIT format are given in terms of sample
numbers (16kHz sample rate), whereas the edited output file is in 
\HTK\ format in which all times are in absolute 100ns units.  

As well as the Replace command, there is
also a {\it Merge} command \texttt{ME}\index{me@\texttt{ME} command}.  This command is used 
to replace a sequence of labels by a single label.  \index{labels!merging}
For example, the following commands would merge the closure and release
labels in the previous TIMIT transcription into single labels
\begin{verbatim}
    ME b bcl b
    ME d dcl dh
    ME t tcl t
\end{verbatim}
As shown by this example, the label used for the merged sequence can be
the same as occurs in the original but some care is needed since
\htool{HLEd} commands are normally applied in sequence.   Thus, a command
on line $n$ is applied to the label sequence that remains after the
commands on lines 1 to $n-1$ have been applied.

There is one exception to the above rule of sequential edit command
application.  The {\it Change} command \index{labels!changing}
\texttt{CH}\index{ch@\texttt{CH} command} provides for context
sensitive replacement.  However, when a sequence of Change commands occur
in a script, the sequence is applied as a block so that the contexts
which apply for each command are those that existed just prior to the
block being executed. The Change command takes 4 arguments \texttt{X A Y
B} such that every occurrence of label \texttt{Y} in the context of
\texttt{A \_ B} is changed to the label \texttt{X}.  The contexts
\texttt{A} and \texttt{B} refer to sets of labels and  are defined by
separate \textit{Define Context} commands \texttt{DC}\index{dc@\texttt{DC} command}.   
The \texttt{CH} and
\texttt{DC} commands are primarily used for creating context sensitive
labels.  For example, suppose that a set of context-dependent phoneme
models are needed for TIMIT.  Rather than treat all possible contexts
separately and build separate triphones for each  (see below), the
possible contexts  will be grouped into just 5 broad classes: C
(consonant), V (vowel), N (nasal), L (liquid) and S (silence). The goal
then is to translate a label sequence such as \texttt{sil b ah t iy n ...}
into \texttt{sil+C S-b+V C-ah+C V-t+V C-iy+N V-n+ ...} where the
\texttt{-} and \texttt{+} symbols within a label are recognised by 
\HTK\ as defining the left and right context, respectively. To perform this
transformation, it is necessary to firstly use \texttt{DC} commands to
define the 5 contexts, that is
\begin{verbatim}
    DC V iy ah ae eh ix ... 
    DC C t k d k g dh ... 
    DC L l r w j ...
    DC N n m ng ...
    DC S h# #h epi ...
\end{verbatim}
Having defined the required contexts, a change command must be
written for each context dependent triphone, that is
\begin{verbatim}
    CH V-ah+V V ah V
    CH V-ah+C V ah C
    CH V-ah+N V ah N
    CH V-ah+L V ah L
     ...
     etc
\end{verbatim}
This script will, of course, be rather long (25 $\times$ number of
phonemes) but it can easily be generated automaticaly 
by a simple program or shell script.

The previous example shows how to transform a set of phonemes into a
context dependent set in which the contexts are user-defined.
\index{labels!context dependent} For
convenience, \htool{HLEd} provides a set of global transformation commands
for converting phonemic transcriptions to conventional left or right
biphones, or full triphones.  For example, a script containing the single
{\it Triphone Conversion} command \texttt{TC}\index{tc@\texttt{TC} command} will convert phoneme 
files to regular
triphones.  As an illustration, applying the \texttt{TC} command to a file
containing the sequence \texttt{sil b ah t iy n ...} would  give the
transformed sequence 
\texttt{sil+b sil-b+ah b-ah+t ah-t+iy t-iy+n iy-n+ ...}. Notice that the
first and last phonemes in the sequence  cannot be transformed in the
normal way.  Hence, the left-most and right-most contexts of these
start and end phonemes can be specified
explicitly as arguments to the \texttt{TC} commands if required.  For
example, the command \texttt{TC \# \#} would give the sequence
\texttt{\#-sil+b sil-b+ah b-ah+t ah-t+iy t-iy+n iy-n+ ... +\#}.  
Also, the contexts at pauses and word boundaries can be blocked
using the
\texttt{WB} command.   For example, if \texttt{WB sp} was executed, the
effect of a subsequent \texttt{TC} command on the sequence 
\texttt{sil b ah t sp iy n ...} would be to give the sequence 
\texttt{sil+b sil-b+ah b-ah+t ah-t sp iy+n iy-n+ ...}, 
where \texttt{sp} represents a short pause. Conversely, the \texttt{NB} 
command can be used to ignore a label as far as context is concerned.  For example,
if  \texttt{NB sp} was executed, the
effect of a subsequent \texttt{TC} command on the sequence 
\texttt{sil b ah t sp iy n ...} would be to give the sequence 
\texttt{sil+b sil-b+ah b-ah+t ah-t+iy sp t-iy+n iy-n+ ...}.

When processing \HTK\ format label files with multiple levels, only the 
level 1 (i.e. left-most) labels are affected.\index{labels!moving level}  To process a higher
level, the \textit{Move Level} command \texttt{ML}\index{ml@\texttt{ML} command} should be used.
For example, in the script
\begin{verbatim}
    ML 2
    RE one 1
    RE two 2
    ...
\end{verbatim}
the Replace commands are applied to level 2 which is the first level
above the basic level.  The command \texttt{ML 1} returns to the 
base level.  A complete level can be deleted by the 
\textit{Delete Level} command \texttt{DL}.  This command can
be given a numeric argument to delete a specific level or
with no argument, the current level is deleted.
Multiple levels can also be split into single level alternatives
by using the \textit{Split Level} command \texttt{SL}.

When processing \HTK\ format files with multiple alternatives,
each alternative is processed as though it were a separate file.

Remember also that in addition to the explicit \htool{HLEd} commands,
levels and alternatives can be filtered on input by setting the
configuration variables \texttt{TRANSLEV}\index{translev@\texttt{TRANSLEV}} and 
\texttt{TRANSALT}\index{transalt@\texttt{TRANSALT}} 
(see section~\ref{s:labstruct}).

Finally, it should be noted that most \HTK\ tools require 
all HMMs used in a system to be defined in\index{HMM lists}
a {\it HMM List}.  \htool{HLEd} can be made to automatically generate such
a list as a by-product of editing the label files by using the 
\texttt{-n} option.  For example, the following command would apply
the script \texttt{timit.led} to all files in the directory \texttt{tlabs},
write the converted files to the directory \texttt{hlabs}
and also write out a list of all new labels in the edited
files to \texttt{tlist}.
\begin{verbatim}
    HLEd -n tlist -l hlabs -G TIMIT timit.led tlabs/*
\end{verbatim}
Notice here that the \texttt{-G} option is used to inform \htool{HLEd} 
that the format of the source files is \texttt{TIMIT}. This could also
be indicated by setting the configuration variable 
\texttt{SOURCELABEL}\index{sourcelabel@\texttt{SOURCELABEL}}.

\mysect{Summary}{labelsum}

Table~\href{t:labelcparms} 
lists all of the
configuration parameters  recognised by \htool{HLabel}
along with a brief description.  A missing module name means 
that it is recognised by
more than one module. 

\begin{center}
\begin{tabular}{|p{1.4cm}|p{3.0cm}|p{6.4cm}|} \hline
Module & Name  & Description  \\ \hline
\htool{HLabel} & \texttt{LABELSQUOTE}  & Specify label quote character \\
\htool{HLabel} & \texttt{SOURCELABEL}    & Source label format \\
\htool{HLabel} & \texttt{SOURCERATE}    & Sample period for SCRIBE format \\
\htool{HLabel} & \texttt{STRIPTRIPHONES} & Remove triphone contexts on input \\
\htool{HLabel} & \texttt{TARGETLABEL}    &  Target label format\\
\htool{HLabel} & \texttt{TRANSALT}  & Filter alternatives on input \\
\htool{HLabel} & \texttt{TRANSLEV}  & Filter levels on input \\
\htool{HLabel} & \texttt{V1COMPAT}  & Version 1.5 compatibility mode \\
               & \texttt{TRACE}           & trace control (default=0) \\ \hline
\end{tabular}
\tabcap{labelcparms}{Configuration Parameters used with Labels}
\end{center}


%%% Local Variables: 
%%% mode: latex
%%% TeX-master: "htkbook"
%%% End: