This thesis presents work done during the last ten years on developing five multimodal spoken dialogue systems, and the empirical user studies that have been conducted with them. The dialogue systems have been multimodal, giving information both verbally with animated talking characters and graphically on maps and in text tables. To be able to study a wider rage of user behaviour each new system has been in a new domain and with a new set of interactional abilities. The five system presented in this thesis are: The Waxholm system where users could ask about the boat traffic in the Stockholm archipelago; the Gulan system where people could retrieve information from the Yellow pages of Stockholm; the August system which was a publicly available system where people could get information about the author Strindberg, KTH and Stockholm; the AdAptsystem that allowed users to browse apartments for sale in Stockholm and the Pixie system where users could help ananimated agent to fix things in a visionary apartment publicly available at the Telecom museum in Stockholm. Some of the dialogue systems have been used in controlled experiments in laboratory environments, while others have been placed inpublic environments where members of the general public have interacted with them. All spoken human-computer interactions have been transcribed and analyzed to increase our understanding of how people interact verbally with computers, and to obtain knowledge on how spoken dialogue systems canutilize the regularities found in these interactions. This thesis summarizes the experiences from building these five dialogue systems and presents some of the findings from the analyses of the collected dialogue corpora. / QC 20100611

情報処理学会研究報告音声言語情報処理;2002-SLP-44-23

情報処理学会研究報告. SLP, 音声言語情報処理; 2003-SLP-49-46 第5回音声言語シンポジウム

<p>This study formed part of a larger project that is concerned with the adaptation of a test of cognitive academic language proficiency, the Woodcock Mu&ntilde / oz Language Survey (WMLS). The WMLS has been adapted from English into isiXhosa and the present study is located within the broader study that is concerned with establishing overall equivalence between the two language versions of the WMLS. It was primarily concerned with the Verbal Analogies (VA) scale. Previous research on this scale has demonstrated promising results, but continues to find evidence of some inequivalence. This study aimed to cross-validate previous research on the two language versions of the WMLS and improve on methodological issues by employing matched groups. It drew upon an existing dataset from the larger research project. The study employed a monolingual matched two-group design consisting of 150 mainly English speaking and 149 mainly isiXhosa learners in grades 6 and 7. This study had two sub aims. The first was to investigate item bias by identifying DIF items in the VA scale across the isiXhosa and English by conducting a logistic regression and Mantel-Haenszel procedure. Five items were identified by both techniques as DIF. The second sub aim was to evaluate construct equivalence between the isiXhosa and English versions of the WMLS on the VA scale by conducting a factor analysis on the tests after removal of DIF items. Two factors were requested during the factor analysis. The first factor displayed significant loadings across both language versions and was identified as a stable factor. This was confirmed by the Tucker&rsquo / s Phi and scatter plot. The second factor was stable for the English version but not for the isiXhosa version. The Tucker&rsquo / s phi and scatter plot indicated that this factor is not structurally equivalent across the two language versions</p>

<p>In the majority of the schools in South Africa (SA), learners commence education in English. This English milieu poses a considerable challenge for English second-language speakers. In an attempt to bridge the gap between English as the main medium of instruction and the nine indigenous languages of the country and assist with the implementation of mother-tongue based bilingual education, this study focuses on the cross-validation of a monolingual English test used in the assessment of multilingual or bilingual learners in the South African context. This test, namely the Woodcock Mu&ntilde / oz Language Survey (WMLS), is extensively used in the United States in Additive Bilingual Education in the country. The present study is a substudy of a broader study, in which the original WMLS (American-English version) was adapted into SA English and Xhosa. For this specific sub-study, the researcher was interested in investigating the scalar equivalence of the adapted English version of the Verbal Analogies (VA) subscale of the WMLS across English first-language speakers and Xhosa first-language speakers. This was achieved by utilising differential item functioning (DIF) and construct bias statistical techniques. The Mantel-Haenszel DIF detection method was employed to detect DIF, while construct equivalence was examined by means of exploratory factor analysis (EFA) utilising an a priori two-factor structure. The Tucker&rsquo / s phi coefficient was used to assess the congruence of the construct across the two language groups</p>

"Resumo/Abstract", "palavras-chave/keywords" nos cap?tulos trabalho. / Submitted by Jos? Henrique Henrique (jose.neves@ufvjm.edu.br) on 2017-03-27T20:52:05Z No. of bitstreams: 2 license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) fernanda_gandra_oliveira.pdf: 3187185 bytes, checksum: 690bd6e7967bbab89412d221de93c88a (MD5) / Approved for entry into archive by Rodrigo Martins Cruz (rodrigo.cruz@ufvjm.edu.br) on 2017-04-24T14:26:09Z (GMT) No. of bitstreams: 2 license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) fernanda_gandra_oliveira.pdf: 3187185 bytes, checksum: 690bd6e7967bbab89412d221de93c88a (MD5) / Made available in DSpace on 2017-04-24T14:26:09Z (GMT). No. of bitstreams: 2 license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) fernanda_gandra_oliveira.pdf: 3187185 bytes, checksum: 690bd6e7967bbab89412d221de93c88a (MD5) Previous issue date: 2016 / Coordena??o de Aperfei?oamento de Pessoal de N?vel Superior (CAPES) / Funda??o de Amparo ? Pesquisa do Estado de Minas Gerais (FAPEMIG) / O comp?sito nFZV-CAG foi empregado para a ozoniza??o catal?tica heterog?nea da NMS em meio aquoso. O oz?nio foi gerado por uma central geradora de oz?nio IPABRAS, alimentada com ar. Foram avaliados diferentes processos catal?ticos como O3, CAG, O3-CAG, O3-nFZV- CAG, e a varia??o da concentra??o das nFZV para remo??o da NMS. Os resultados mostraram que a combina??o do O3-nFZV-CAG foi muito eficiente levando ? mineraliza??o de aproximadamente 70% da NMS em 120 min de rea??o. Tal efici?ncia pode estar atribu?da ao processo de eletr?lise, em que o Fe0 origina Fe2+ levando ? produ??o do radical hidroxila que ? altamente oxidante, levando a destrui??o do contaminante. As rea??es seguiram a cin?tica pseudo-primeira ordem para remo??o do f?rmaco. Ap?s realizada a coleta as amostras foram submetidas a an?lise qu?mica, empregando-se as t?cnicas de espectrofotometria de UV-Vis, cromatografia l?quida de alta efici?ncia (CLAE) e a determina??o da demanda qu?mica de oxig?nio (DQO) tamb?m foi realizada. / As nanopart?culas de FZV imobilizadas sobre a superf?cie do carv?o ativado (nFZV-CAG) foram sintetizadas e caracterizadas para remo??o do f?rmaco Nimesulida (NMS) em sistemas aquosos. Os estudos foram realizados em bateladas com agita??o de 250 rpm durante 120 minutos, onde foram avaliadas a efici?ncia, concentra??o (20, 25 e 30% do comp?sito) e diferentes dosagens (0,1 a 10g) de nFZV-CAG. Os resultados mostraram que a dosagem de 10 g de 20%nFZV-CAG removeu cerca de 80% da NMS 50 mg L-1 em apenas 30 minutos de rea??o, e atingindo 100% em 120 minutos. Foi observada ainda uma remo??o de 80% da DQO ao final da rea??o. As velocidades de rea??o aumentaram na medida em que foram aumentadas as dosagens do comp?sito, o que seria esperado. Por?m, com o aumento da concentra??o (%) de nFZV n?o houve aumento na velocidade das rea??es. As rea??es seguiram uma cin?tica de pseudo-primeira ordem em rela??o ? remo??o da NMS. Ap?s realizada a coleta, as amostras foram submetidas a an?lise qu?mica, empregando-se as t?cnicas de espectrofotometria de UV- VIS, cromatografia l?quida de alta efici?ncia (CLAE) e a determina??o da demanda qu?mica de oxig?nio (DQO) tamb?m foi realizada. Para caracteriza??o do comp?sito foram empregadas as t?cnicas Microscopia Eletr?nica de Varredura acoplada ? Espectrometria de Energia Dispersiva de Raios-X (MEV-EDS), que mostraram claramente a presen?a da nanopart?culas sobre a superf?cie do carv?o, e analise de superf?cie do nFZV-CAG e do CAG tamb?m foram realizadas. / Disserta??o (Mestrado) ? Programa de P?s-Gradua??o em Qu?mica, Universidade Federal dos Vales do Jequitinhonha e Mucuri, 2016. / The FZV nanoparticles immobilized onto the surface of the activated charcoal (nFZV-CAG) were synthesized and characterized for the removal of the pharmaceutical Nimesulide (NMS) in aqueous systems. The studies were performed in batch mode with stirring speeds of 250 rpm during 120 minutes, so that it was possible to evaluate efficiency, concentration (20, 25 and 30% of the composite) and different dosages (0,1 to 10g) of nFZV-CAG. The results showed that the dosage of 10g of 20%nFZV-CAG removed approximately 80% of NMS 50 mg L-1 in just 30 minutes of reaction, and reaching 100% in 120 minutes. It was still observed a removal of 80% of the COD (Chemical Oxygen Demand) at the end of the reaction. The reaction rates increased with the dosage of the composite, which was expected. Though, increasing the concentration (%) of nFZV did not result in higher reaction rates. The reactions followed a pseudo-first order kinetics for the removal of NMS. After the samples were collected, they were submitted to chemical analysis, employing the techniques of UV-VIS spectroscopy, high performance liquid chromatography (HPLC) and chemical oxygen demand (COD). In order to characterize the composite, the following techniques were used: Scanning Electron Microscopy (SEM) coupled with Energy Dispersive x-Ray spectroscopy (EDS), which showed clearly the presence of nanoparticles onto the charcoal surface. Analysis of the surface of nFZV-CAG were also performed. / The composite nFZV-CAG was used for the catalytic ozonation of NMS in aqueous media. The ozone was generated by an ozone generator IPABRAS, fuelled with air. Different catalytic processes were evaluated such as O3, CAG, O3-CAG, O3-nFZV-CAG, and the variation of the nFZV concentration for the removal of NMS. The results showed that the combination of O3- nFZV-CAG was very efficient, leading to the mineralization of approximately 70% of NMS in 120 minutes of reaction. Such efficiency can be attributed to the electrolysis process, in which the Fe0 generates Fe2+ which generates hydroxyl radicals that are highly oxidant, leading to the destruction of the contaminant. The reactions followed the pseudo-first order kinetics for the removal of the pharmaceutical. After the samples were collected, they were submitted to chemical analysis such as, UV-VIS spectroscopy, high performance liquid chromatography (HPLC) and chemical oxygen demand (COD).

"Measure twice and cut once" MSc. Kunc´s super8 format narrative fabrication in 80´s. This dissertation presents amateurish authorial films of MSc .Vladimír Kunc on Super 8 mm film format with emphasis on film editing as an reflection of time in the moment of its origin in the 80ties. I will compare the phenomenon of applied magnetic sound of Dr.Rentz for Kunz image montages with a sound of composition of authorial films of The Wizzard of Oz, who has chosen adventurous way of audio tape to his films on Super 8mm film format in the same decade. The aspect of authorial sound montage is looked through the contemporary technological and distributive possibilities.The text also presents analysis of expressional means of Super 8 mm films and the way of their using within particular authorial approaches-model authorial subjects. This dissertation presents study of narrative films which during the filming and editing look for not only notice value but also the way of narration.The films which are looking not only for message but also its form in the context particular period.

Language is the principal medium for ideas, while dialogue is the most natural and effective way for humans to interact with and access information from machines. Natural language generation (NLG) is a critical component of spoken dialogue and it has a significant impact on usability and perceived quality. Many commonly used NLG systems employ rules and heuristics, which tend to generate inflexible and stylised responses without the natural variation of human language. However, the frequent repetition of identical output forms can quickly make dialogue become tedious for most real-world users. Additionally, these rules and heuristics are not scalable and hence not trivially extensible to other domains or languages. A statistical approach to language generation can learn language decisions directly from data without relying on hand-coded rules or heuristics, which brings scalability and flexibility to NLG. Statistical models also provide an opportunity to learn in-domain human colloquialisms and cross-domain model adaptations. A robust and quasi-supervised NLG model is proposed in this thesis. The model leverages a Recurrent Neural Network (RNN)-based surface realiser and a gating mechanism applied to input semantics. The model is motivated by the Long-Short Term Memory (LSTM) network. The RNN-based surface realiser and gating mechanism use a neural network to learn end-to-end language generation decisions from input dialogue act and sentence pairs; it also integrates sentence planning and surface realisation into a single optimisation problem. The single optimisation not only bypasses the costly intermediate linguistic annotations but also generates more natural and human-like responses. Furthermore, a domain adaptation study shows that the proposed model can be readily adapted and extended to new dialogue domains via a proposed recipe. Continuing the success of end-to-end learning, the second part of the thesis speculates on building an end-to-end dialogue system by framing it as a conditional generation problem. The proposed model encapsulates a belief tracker with a minimal state representation and a generator that takes the dialogue context to produce responses. These features suggest comprehension and fast learning. The proposed model is capable of understanding requests and accomplishing tasks after training on only a few hundred human-human dialogues. A complementary Wizard-of-Oz data collection method is also introduced to facilitate the collection of human-human conversations from online workers. The results demonstrate that the proposed model can talk to human judges naturally, without any difficulty, for a sample application domain. In addition, the results also suggest that the introduction of a stochastic latent variable can help the system model intrinsic variation in communicative intention much better.

In this work is designed programme unit for acquisition analog acoustic signlas from sensors. There is used accurate A/D converter ADS1287 by the company Texas Iinstrument with resolution of 24 bits to digitizing these signals. There is used 32-bit microprocessor AT91SAM7S64 by the company Atmel to control this A/D converter and sending digitized data to PC. This microprocessor has implemented USB interface. By force of developmental programme units whit microprocessor and A/D converter is produced programme for microprocessor in developmental setting IAR Embedded Workbench IDE 5.0 and simple aplication for PC in setting Borland C++ Builder. Both of those programs are in language C++. Rate of sampling is 26 kHz. On the basis is realized programme unit USB-ADC whit microprocessor and A/D converter.

This work deals with using modern active functional blocks in electronic circuits. These blocks have better features than classical operational amplifiers and are characterized especially by working in current mode. Therefore these blocks can be used in higher frequencies. Using those blocks it is possible to design e.g. electronically variable-frequency filters of types low-pass filter, high-pass filter, band-pass filter, etc. Then it is also possible to easily realize electronically variable-frequency oscillators. This work is focused on some of these blocks, their basic characteristics and principles. Also three oscillators with current conveyors of second generation are designed. Its made simulation, sensitivity and tolerance analysis and magnitude stabilization. Everything is performed in OrCAD PSpice program. These made oscillators are electronic tunable in the band frequency ones of MHz.