Báo cáo khoa học: "Reconstructing false start errors in spontaneous speech text" ppt

... utterances used in training. In some experiments (#1 and 2) wetrained with the entire training set3, 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 concurrentlymaintaining or improving precision.1.1 Error classes in spontaneous speech Common ... false start errors in spontaneous speech textErin FitzgeraldJohns Hopkins UniversityBaltimore, MD, USAerinf@jhu.eduKeith HallGoogle, Inc.Zurich, Switzerlandkbhall@google.comFrederick JelinekJohns...

Tài liệu Báo cáo khoa học: "Resumption strategies for an in-vehicle dialogue system" ppt

... work-load.dialogue-inducedpossible driving-induceddriving-inducedlow workloadPage 1Figure 7: Workload while driver is resuming to theinterview domain.dialogue-inducedpossible driving-induceddriving-inducedlow ... driving task is demanding)and possibly driving-induced when only IDIS isindicating high workload (since IDIS admittedlyis indicating that the driving task is demandingbut the TDT indicates that ... respectively.dialogue-inducedpossibly driving-induceddriving-inducedlow workloadPage 1Figure 5: Workload while the driver is interruptingan interview topic.dialogue-inducedpossibly driving-induceddriving-inducedlow...

Tài liệu Báo cáo khoa học: "Non-Projective Dependency Parsing in Expected Linear Time" pptx

... 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 ofthe new system on naturally occurring data is in fact ... trees by swappingthe order of words in the input. Addingthe swapping operation changes the timecomplexity for deterministic parsing fromlinear to quadratic in the worst case, butempirical estimates ... approach we will take in the experimental evaluation in Section 4.3 Transitions for Dependency ParsingHaving deﬁned the set of conﬁgurations, includinginitial and terminal conﬁgurations, we...

Tài liệu Báo cáo khoa học: "Identifying Agreement and Disagreement in Conversational Speech: Use of Bayesian Networks to Model Pragmatic Dependencies" docx

... single meeting as a test set and the remaining dataas training material; for this experiment, we usedthe 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 anddisagreement. By integrating the interpretation ofthe pragmatic ... Entropy RankingWe view the problem as an instance of statisti-cal ranking, a general machine learning paradigmused for example in statistical parsing (Collins,2000) and question answering (Ravichandran...

Tài liệu Báo cáo khoa học: "Constraints over Lambda-Structures in Semantic Under specification" pptx

... (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 binding relations ... antecedents are terminal nodes; (6) there are no cyclic link chains; (7) if a link chain ends at a variable then each anaphor in the chain must be dominated by the binder of that variable....

Báo cáo khoa học: "Importance of linguistic constraints in statistical dependency parsing" ppt

... 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 children in the ... integer pro-gram as McDonald et al. (2006) has formulated dependency parsing as MST problem. All the linguistic constraints are encoded as constraints while solving the integer program. In...

Báo cáo khoa học: "Vocabulary Decomposition for Estonian Open Vocabulary Speech Recognition" ppt

... “Morfessor Baseline” is described in detail in (Creutz et al., 2007). An unsupervisedmachine learning algorithm is used to discover themorph set of the language in question, using mini-mum description ... provided here as a baseline forthe alternative approaches. The alternatives testedhere are all based on decomposing the in- vocabularywords, OOV words, or both, in LM training data intosequences of ... language in questionhas signiﬁcant orthographic irregularity.As can be seen, both types of morphological com-plexity increase the amount of distinct word forms,resulting in an increase in the...

Báo cáo khoa học: "Learning Intonation Rules for Concept to Speech Generation" pptx

... 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 system, CTS ... used in our intonation model are mainly extracted from this final FD. The intonation theory proposed in (Pierre- humbert, 1980) is used to describe the intona- tion structure. Based on her intonation...

Báo cáo khoa học: "USE OF HERRISTIC KNOWLEDGE IN CHINNESE LANGUAGE ANALYSIS" ppt

... 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 determine all the ... by the following principles: a) Kind of fragment Some kinds of fragments, for example, com- pound verbs involving "~", occur more often than others and are accordingly given higher...

Tài liệu Báo cáo khoa học: "How Are Spelling Errors Generated and Corrected? " docx

... ours in that theydid not directly log the input in pinyin (RomanizedChinese by which native characters are input), butthe input pinyin sequences are recovered from theChinese words in the ... bottom.”)3.2 Task interfaceFor logging the keystrokes including the use ofbackspaces, we designed an original interface for thetext boxes in the MTurk task. In order to simplifythe interpretation ... words from appearing in thelist of candidates in their native scripts, thereby pre-venting them from being entered altogether. Finally,we can collect a large amount of spelling errors andtheir...

Xem thêm