Information Technology Reference
In-Depth Information
Ta b l e 3 .
Response times in different test scenarios
i
ii
iii
iv
v
0.08 s 0.28 s 2.37 s 2.67 s 9.06 s
base
action
pre-select
0.08 s 0.24 s 2.10 s 2.29 s 7.15 s
entity
pre-select
0.06 s 0.19 s 2.01 s 2.16 s 7.41 s
parameter
pre-select.
0.09 s 0.19 s 1.06 s 1.20 s 4.05 s
action +
entity
0.05 s 0.16 s 1.70 s 1.85 s 6.07 s
entity +
parameter
0.05 s 0.13 s 0.99 s 1.10 s 3.75 s
action +
parameter
0.09 s 0.13 s 0.71 s 0.83 s 2.52 s
full com-
bination
0.07 s 0.10 s 0.68 s 0.76 s 2.35 s
while the depth of the search tree for utterance (v) is 7. Note that utterance (iv) in-
volves two distinct search trees, since it contains two independent verb phrases which
are interpreted separately.
The five utterances were tested with the following versions of the system. First, we
used the base system as described in Section 3, it does not include any explicit perfor-
mance improvements speed-wise. The first row of Table 3 shows the performance of
thebasesystem.
Improvements.
Second, we considered systems incorporating different pre-selection
methods. For each interpretation step (interpreting action, entity and parameter), we
can pre-select the candidates that may be considered by the appropriate interpretation
action. This can lead to considerably lower branching factors.
The pre-selection process for
interpret action
involves two criteria: synonym and
parameter count. This means that candidates are eliminated from the list if the spoken
verb is not one of the candidates' synonyms or if the number of parameters the candidate
provides is lower than the number of spoken objects. This is due to the fact that we want
every spoken object to be assigned to a parameter slot, so we only have to consider skills
that provide a sufficient amount of parameter slots. If we would also consider skills with
fewer parameters, we would have to drop parts of the user's utterance. One could argue
that reducing the set of available skills is a restriction from a theoretical point of view.
However, ignoring elements that where uttered could easily frustrate the user. Hence,
the restriction only has little practical relevance. The second row of Table 3 illustrates
the performance of the base system plus
action pre-selection
.
Entities are pre-selected just by checking whether the spoken object is one of the
entity's synonyms. The third row of Table 3 shows the response times including the
base system plus
entity pre-selection
.
Pre-selecting parameters involves checking the attributes and the preposition of the
corresponding candidate. Hence, the attributes of the parameter slot have to be a subset
of the entities attributes, and if a preposition was provided along with the spoken object
or entity, respectively, then it has to match the preposition required by the parameter.
The fourth row of Table 3 lists response times of the base system plus
parameter pre-
selection
. Rows five, six and seven illustrate the performance of different pairs of the
