Báo cáo khoa học: "SPEECH DIALOGUE WITH FACIAL DISPLAYS: MULTIMODAL HUMAN-COMPUTER CONVERSATION" docx

... flexibility of human-computer dialogue and reducing the men- tal barriers that many users associate with computer systems. Research Overview of Multimodal Dialogues Multimodal dialogues that ... SPEECH DIALOGUE WITH FACIAL DISPLAYS: MULTIMODAL HUMAN-COMPUTER CONVERSATION Katashi Nagao and Akikazu Takeuchi Sony Computer ... Speech Dialogue and Facial Animation We have developed an experimental system that integrates speech dialogue and facial animation to investigate the effects of human-like behavior in human-computer...

Báo cáo khoa học: "Chinese Segmentation with a Word-Based Perceptron Algorithm" docx

... sentenceis turned into the raw input form, and then decoded with the current parameter vector. The output seg-mented sentence is compared with the original train-ing example. If the output is incorrect, ... generatedcandidates can be combined with the next incom-ing character to generate new candidates. After thelast character is processed, the decoder returns thecandidate with the best score in the source ... partialcandidates ending with the same bigram, the bestone will also be in the best ﬁnal candidate. Thedecoder can be optimized accordingly: when an in-coming character is combined with candidate itemsas...

Tài liệu Báo cáo khoa học: "Context Management with Topics for Spoken Dialogue Systems" pptx

... we discuss the use of discourse context in spoken dialogue systems and argue that the knowledge of the domain, modelled with the help of dialogue topics is important in maintaining robust- ... topic identification has been applied to dialogue move (dialogue act) recognition (Garner, 1997). Our goal is to build a context model for a spoken dialogue system, and we emphasise especially ... utterance topic is an important task of spoken dialogue systems. 3 The Topic Model In AI-based dialogue modelling, topics are associated with a particular discourse entity, focus, which...

Báo cáo khoa học: "Speech emotion recognition with TGI" pot

... inference algo-rithm. We have extended this part of thealgorithm with various edit costs to pe-nalise more important features with higheredit costs for being outside the interval,which tree ... branches. The motivation behindis that acoustically some emotions are transmitted with segmental features and others with prosodic,e.g. prosody can be prioritised over segmental fea-tures or vice ... device. The motivation behindis that acoustically some emotions are transmitted with segmental features and others with prosodic(Barra, et al., 1993). A coefﬁcient of 1.5 on theprosodic branches...

Tài liệu Báo cáo khoa học: Amprenavir complexes with HIV-1 protease and its drug-resistant mutants altering hydrophobic clusters docx

... PR with the V32I mutation; PRI50V, PR with the I50V mutation; PRI54M, PR with the I54M mutation; PRI54V, PR with the I54V mutation; PRI84V, PR with the I84V mutation; PRL90M, PR with ... compared with PRWT–SQV. (B) PRI50V–APV compared with PRI50V–SQV. (C)PRI54M–APV compared with PRI54M–SQV. (D) PRI54V–APV compared with PRI54V–SQV. (E) PRI84V–APV compared with PRI84V–SQV. ... the complexes with both inhibitors.DiscussionWe have described an atomic resolution crystal struc-ture of PRWT with APV, and analyzed structuralchanges in the APV complexes with mutants PRV32I,PRI50V,PRI54M,PR154V,PRI84Vand...

Tài liệu Báo cáo khoa học: A tyrosinase with an abnormally high tyrosine hydroxylase/dopa oxidase ratio Role of the seventh histidine and accessibility to the active site docx

... R3. The PCR product with a size of  1.6 kbwas digested with BamHI to obtain a fragment betweenthe two copper-binding site coding regions and ligatedto pBlueScript pKSII(+) with T4 DNA ligase ... R3-1501–than R3-0337–. This pattern agrees with the presence of twodifferent enzymes with overlapping activities.Purification of two enzymes with differentaffinities for monophenols and o-diphenolsSupernatants ... differentforms during the cycles: met, resting tyrosinase with Cu(II); oxy,oxygenated form with peroxide bound to Cu(II); deoxy, reducedCu(I) transient form with high affinity for oxygen. The efficiency forboth...

Tài liệu Báo cáo khoa học: "Semantic Parsing with Bayesian Tree Transducers" doc

... a pattern with which to expandit and then filling out that pattern with words drawnfrom a unigram distribution.This kind of coupled generative process canbe naturally formalized with tree ... gen-erate MR and NL associated with MR rules m -v (top).Transducer rule 2 selects MR rule r from the MR gram-mar. Rule 3 simultaneously writes the MR associated with rule m and chooses an NL ... Ψ(r:s(r)=tˆα(r)).493The parameters of q(θt) are defined with respectto q(xi) and the parameters of q(xi) with respectto the parameters of q(θt). q(xi) can be computedefficiently...

Tài liệu Báo cáo khoa học: "Word Alignment with Synonym Regularization" doc

... the SRH.We show the performance of GIZA++ and HM-BiTAM with the SRH in the lines entitled with SRH” in Table 1. The GIZA++ and HM-BiTAM with the SRH slightly outperformed the standardGIZA++ ... alignment model. We formulate a syn-onym pair generative model with a topic variableand use this model as a regularization term with abilingual word alignment model. The topic vari-able in ... meanings of synonyms. We extendHM-BiTAM, which is a HMM-based word align-ment model with a latent topic, with a novel syn-onym pair generative model. We applied the pro-posed method to an...

Tài liệu Báo cáo khoa học: "Coreference Resolution with Reconcile" pdf

... infrastructure to sup-port the creation of, experimentation with, andevaluation of coreference resolvers. Reconcilewas designed with the following seven desideratain mind:• implement the ... present).The basic architecture of the system includesﬁve major steps. Starting with a corpus of docu-ments together with a manually annotated corefer-ence resolution answer key1, Reconcile performs1Only ... original Javacode. Only about 15% of the code is concerned with running existing components in the prepro-cessing step, while the rest deals with NP extrac-tion, implementations of features, clustering...

Tài liệu Báo cáo khoa học: "Disentangling Chat with Local Coherence Models" pdf

...

Xem thêm