... utterances used in
training. In some experiments (#1 and 2) we
trained with the entire training set
3
, including sen-
tences without speaker errors, and in others (#3-6)
we trained only on those ... al., 2004),
aiming to improve overall recall – especially of
false start or non-copy errors – while concurrently
maintaining or improving precision.
1.1 Error classes in spontaneo...
... work-
load.
dialogue-
induced
possible
driving-
induced
driving-
induced
low
workload
Page 1
Figure 7: Workload while driver is resuming to the
interview domain.
dialogue-
induced
possible
driving-
induced
driving-
induced
low ... driving task is demanding)
and possibly driving-induced when only IDIS is
indicating high workload (since IDIS admittedly
is indicating that the driving task is...
... quadratic in the worst case but still linear
in the best case. In Section 4, we present experi-
ments indicating that the expected running time of
the new system on naturally occurring data is in
fact ... trees by swapping
the order of words in the input. Adding
the swapping operation changes the time
complexity for deterministic parsing from
linear to quadratic in the worst case,...
... single meeting as a test set and the remaining data
as training material; for this experiment, we used
the same training set as (Hillard et al., 2003). Per-
formance is reported in Table 6. In ... several empirical analyses in or-
der to determine to what extent contextual informa-
tion helps in discriminating between agreement and
disagreement. By integrating the interpretation of
th...
... (Niehren and Koller, 1998) if
binding and linking relations in A-structures are
ignored.
For the interaction of binding with parallelism
we follow the basic idea that binding relations
should be ... variable-capturing when instan-
tiating underspecified scope representations. In
principle, capturing may occur in all formalisms
for structural underspecification which repre-
sent...
... handling these
kind of linguistic constraints using integer pro-
gramming approaches (Riedel et al., 2006; Bha-
rati et al., 2008). In these approaches dependency
parsing is formulated as solving ... in-
corporating linguistic constraints in statistical
dependency parsing. We consider a simple
linguistic constraint that a verb should not
have multiple subjects/objects as its childr...
... “Morfessor Baseline” is described
in detail in (Creutz et al., 2007). An unsupervised
machine learning algorithm is used to discover the
morph set of the language in question, using mini-
mum description ... provided here as a baseline for
the alternative approaches. The alternatives tested
here are all based on decomposing the in- vocabulary
words, OOV words, or both, in LM training...
...
lexical information, which are produced during
the generation process, are kept in a final Func-
tional Description (FD), before linearizing the
syntactic structure into a linear string. The ... (McKeown et al., 1997) .
Integrating language generation with speech
synthesis within a Concept-to -Speech (CTS)
system not only brings the individual benefits
of each; as an integrated s...
...
In another exa~ole, in English, "-ing" is
used to indicate a participle, or "-ed" can be
used to distinguish passive mode from active. In
Chinese, there is nothing to indicate ... next.
The resulting combination (fl,f3) is correct.
Fig.3 shows the parsing result obtained by
computer in our preprocessing subsystem.
4. PRIORITY
In the preprocessing, we deter...
... ours in that they
did not directly log the input in pinyin (Romanized
Chinese by which native characters are input), but
the input pinyin sequences are recovered from the
Chinese words in the ... bottom.”)
3.2 Task interface
For logging the keystrokes including the use of
backspaces, we designed an original interface for the
text boxes in the MTurk task. In order to simplify
the int...