Báo cáo khoa học: "Enriching spoken language translation with dialog acts" ppt

... in a for- eign language. In the next section, we describe our study with the Spoken Dutch Corpus, and il- lustrate the performed data extraction strategies. 38 4 A study with the Spoken Dutch Corpus The ... to the syllabic inventory of the spoken language. Besides the inventory of spoken sylla- bles, we are interested in the distribution of sylla- bles across the language. 3.1...

Ngày tải lên: 20/02/2014, 04:20

... Speech Translation Evalua- tion. In Processings of the International Workshop on Spoken Language Translation 2005. Philipp Koehn, Franz J. Och, and Daniel Marcu. 2003. Statistical Phrase-based Translation. ... probability of the translation "in order to avoid" is higher than that of the translation "can we avoid". Thus, our method selects the former as t...

Ngày tải lên: 20/02/2014, 04:20

... a large search space (as in modern statistical parsing and machine translation) . It is a major effort to break into the field with a new system, and modifying existing systems—even in a conceptually simple ... e.g., word(loves,q,r) with value p(portion of speech signal | loves). To find the probability of the best parse instead of the total probability of all parses, simply change the val...

Ngày tải lên: 20/02/2014, 16:20

... features. We achieved 65.5% translation accuracy in a German-English translation task vs. 53.2% with IBM Model 4. 1 Introduction Recent research in machine translation challenges us with the exciting problem ... translation as a subtask of machine translation. This en- ables us to build a dedicated noun phrase translation subsystem that improves over the currently best general...

Ngày tải lên: 20/02/2014, 16:20

... labeled by source language words c i (like “y ˇ u”) or variables from a set X = {x 1 , x 2 , . . .}; rhs(r) is expressed in the target language dependency structure with words e j (like with ) and ... Sharon))” h f 1 NP 0,3 “(Bush) ⊔ (with (Sharon))” NPB 0,1 “Bush” B ` ush ´ ı h f 0 CC 1,2 with y ˇ u VP 1,5 “held Sharon))” PP 1,3 with (Sharon)” P 1,2 with NPB 2,3 “Sharon” Sh ¯ al...

Ngày tải lên: 07/03/2014, 22:20

... indepen- dently in training. 3 Discriminative Language Model with Pseudo-Negative samples We propose a novel discriminative language model; a Discriminative Language Model with Pseudo- Negative samples (DLM-PN). ... achieves high perfor- mance with very few features. We then examined the effect of PKI. Table 3 shows the results of the classifier with rd order polynomial kernel bo...

Ngày tải lên: 08/03/2014, 02:21

... this difficulty is to factor the language constraints such that speech decoding is accomplished using a covering grammar with a smaller FSN representation and language decoding is accomplished ... that the size of the 299 grammar, used for maximum likelihood decoding, is kept within reasonable bounds without a loss in the performance. In this paper we propose an approach in whic...

Ngày tải lên: 08/03/2014, 07:20

... interfaced with an existing, implemented transfer com- ponent (Verbmobi1), by translating f-structures into a term language, and by interfacing f- structure representations with an existing ... terface f-structure representations directly with the named semantic-based transfer approach. Third, we compare the two approaches with each other, and with co-description and restr...

Ngày tải lên: 20/02/2014, 18:20

... the time at which the document was created. With the introduction of the TimeBank corpus (Pustejovsky et al., 2003), a set of documents an- notated with temporal information, it became pos- sible ... introduction of the Time- Bank” may OVERLAP with “Machine learning be- coming possible” because “introduction” can be understood as a process that is not finished with the release of the d...

Ngày tải lên: 08/03/2014, 00:20

... where each translation lattice node is augmented with syntactic language model state ˜τ t . In phrase-based translation, many translation lat- tice nodes represent multi-word target language phrases. ... oracle-best translations, but did not improve actual translation results. Post and Gildea (2009) use tree substitution grammar parsing for language modeling, but do not use thi...

Ngày tải lên: 20/02/2014, 04:20