In section 4.3, we described a procedure to
counter in speech data. This technique can also
be used for the procedure for dealing with filled-pauses in spontaneous
speech. We therefore propose two techniques for this
procedure.
Filled-pauses are already known as phone strings
(e.g. ``well'' is /W/ /EH/ /L/ ), and recorded in a
word dictionary. And this method skips
filled-pauses like a . Additionally, we
use the unigram probabilities of filled-pauses as
the penalty.
For example, let's assume that the speech data is as follows.
`` Do you already have well a registration form ? ''
In this case, ``well'' is a filled-pause. The trigram probabilities are calculated as follows.
``do''
``you''
``do''
``already''
``do''
``you''
``well''
``have''
``you''
``already''
``a''
``already''
``have''
``registration''
``have''
``a''
In this equation,
``well''
is the penalty and
``have''
``you''
``already''
means to skip filled-pauses.
As filled-pauses are considered to be sequences of phones, this technique skips phones strings as filled-pauses, we also use a penalty for the phone trigram probabilities. This technique is also used for the procedure for dealing with hesitations, retractions and out-of-vocabulary words .
For example, let's assume the speech data is as follows.
The trigram probabilities are calculated as follows.
``do''
``you''
``do''
``already''
``do''
``you''
``have''
``you''
``already''
``a''
``already''
``have''
``registration''
``have''
``a''
In this equation,
,
and
mean the penalty and
``have''
``you''
``already''
means to skip phone strings.