• Refine Query
  • Source
  • Publication year
  • to
  • Language
  • 71
  • 13
  • 12
  • 10
  • 8
  • 8
  • 8
  • 3
  • 2
  • 1
  • 1
  • 1
  • 1
  • Tagged with
  • 162
  • 55
  • 53
  • 42
  • 24
  • 23
  • 23
  • 22
  • 20
  • 20
  • 19
  • 18
  • 12
  • 11
  • 10
  • About
  • The Global ETD Search service is a free service for researchers to find electronic theses and dissertations. This service is provided by the Networked Digital Library of Theses and Dissertations.
    Our metadata is collected from universities around the world. If you manage a university/consortium/country archive and want to be added, details can be found on the NDLTD website.
31

Computerized Oral Proficiency Test for Japanese: Measuring L2 Speaking Ability with ASR Technology

Matsushita, Hitokazu 08 July 2011 (has links) (PDF)
Developing a time- and cost-efficient method for second language (L2) oral proficiency measurement is one of the research topics that has attracted much attention in recent decades. The purpose of this study is to develop a computerized oral testing system for L2 Japanese using automatic speech recognition (ASR) technology. Two testing methods called elicited imitation (EI) and simulated speech (SS) are proposed to quantify L2 accuracy and fluency via ASR processing. This study also suggests systematic EI item creation leveraging corpus technology and discusses the effectiveness of the test items created through analyses of item difficulty. Further, refinement of the EI grading system is described through a series of statistical investigations. For SS, this study reports the five most influential L2 fluency features identified through machine learning and proposes a method to yield individual SS scores with these features based on previous studies. Lastly, several methods to combine the EI and SS scores are presented to estimate L2 oral proficiency of Japanese.
32

Video Based Automatic Speech Recognition Using Neural Networks

Lin, Alvin 01 December 2020 (has links) (PDF)
Neural network approaches have become popular in the field of automatic speech recognition (ASR). Most ASR methods use audio data to classify words. Lip reading ASR techniques utilize only video data, which compensates for noisy environments where audio may be compromised. A comprehensive approach, including the vetting of datasets and development of a preprocessing chain, to video-based ASR is developed. This approach will be based on neural networks, namely 3D convolutional neural networks (3D-CNN) and Long short-term memory (LSTM). These types of neural networks are designed to take in temporal data such as videos. Various combinations of different neural network architecture and preprocessing techniques are explored. The best performing neural network architecture, a CNN with bidirectional LSTM, compares favorably against recent works on video-based ASR.
33

Economy of Scale of Energy Intensity in Aquifer Storage and Recovery (ASR)

Rapp, Alyson Haley 07 December 2023 (has links) (PDF)
More water utilities are adopting Aquifer Storage and Recovery (ASR) to balance long-term water supply and demand. Due to large implementation and operation costs, ASR projects need to be optimized, particularly for energy use, which is a major operating expense. This study examines the relationships among energy use, recharge, and recovery at two ASR projects in the western United States. The major finding is an economy of scale for recovery processes, but not for gravity-fed recharge processes. The economy of scale found is as follows: the energy intensity recovered decreases with volume. This suggests it is more energy-efficient to recover large volumes of water in one interval instead of recovering smaller volumes at more frequent intervals. The H2Oaks recovery process experienced a 78% decrease in energy intensity from 0 to 50,000 m^3 recovered, while the Sand Hollow site experienced a 43% decrease in energy intensity from 0 to 50,000 m^3 recovered. Statistical analyses of the recovery process showed p values lower than 0.0001, R^2 values between 0.43 and 0.57, and a RMSE value between 0.55 and 2.1, indicating the presence of a moderate correlation between energy and volume. This economy of scale has been observed in multiple instances in water and wastewater treatment. This finding not only has applications to ASR but also all recovery or recharge wells, whether or not they are paired with each other. Furthermore, this study confirms the need for more reliable and accessible energy data to fully understand the implications of the energy–water nexus.
34

Enhancement of Polar WRF atmospheric and surface processes: An annual simulation

Wilson, Aaron Benjamin 23 August 2010 (has links)
No description available.
35

Achieving Automatic Speech Recognition for Swedish using the Kaldi toolkit / Automatisk taligenkänning på svenska med verktyget Kaldi

Mossberg, Zimon January 2016 (has links)
The meager offering of online commercial Swedish Automatic Speech Recognition ser-vices prompts the effort to develop a speech recognizer for Swedish using the open sourcetoolkit Kaldi and publicly available NST speech corpus. Using a previous Kaldi recipeseveral GMM-HMM models are trained and evaluated against commercial options toallow for reasoning of the performance of a customized solution for Automatic SpeechRecognition to that of commercial services. The evaluation takes both accuracy andcomputational speed into consideration. Initial results of the evaluation indicate a sys-tematic bias in the selected test set confirmed by a follow up investigative evaluation.The conclusion is that building a speech recognizer for Swedish using the NST corpusand Kaldi without expert knowledge is feasible but requires further work. / En taligenkännare för svenska utvecklas med målet att utvärdera hur en taligenkännareutvecklad med fritt tillgängliga verktyg står sig mot kommersiella taligenkänningstjänster.Verktyget som används är det öppna källkodsverktyget Kaldi och som träningsdataanvänds det offentligt tillgängliga talkorpuset för svenska från NST. De framtagna mod-ellerna jämförs mot kommersielt tillgängliga tjänster för taligenkänning på svenska.Tidiga resultat i jämförelsen indikerar ett systemiskt jäv i den valda testdata, vilketbekräftas av en uppföljande undersökande utvärdering. Slutsatsen av arbetet är attutsikterna att ta fram en taligenkännare för svenska är goda men kräver omfattandearbete.
36

WebXR Voice Assistant : A comparative study of automatic speech recognition implementation methods in a web-based VR environment

Berglin, Elias January 2022 (has links)
Fully autonomous cars are on the horizon. Knightec wants to enable passengers of the future car to be more productive and entertained with a new web platform. With this platform, Knightec wants to explore different input methods one of which being a voice assistant. A key component in a voice assistant is Automatic Speech Recognition (ASR) and for this task, Knightec had planned to use the new Web Speech API. Their target platform (Oculus Quest 2) does not yet support the Web Speech API and a future implementation could be limited. This thesis conducts a comparative study to find alternatives for running ASR in a web application. The study aimed to compare browser-implemented ASR methods to server implemented methods with Web Speech API as a baseline. The study first conducted a document study to find methods for running ASR tasks inside a web application and then create requirements for method selection. With the requirements, two suitable implementations were found for a browser implementation of ASR. During the final implementation, one of these failed, leaving only one method implemented in the browser. Three ASR methods were chosen for the server implementation, following requirements also set by the document study. To compare the ASR methods a dataset was created with the help of Knightec. The dataset consists of 10 commands, utilizing the voices of six individual employees at Knightec including separate versions, one with and one without background noise for each voice totaling 120 recordings. The dataset was used as a benchmark for each implementation where Word Error Rate (WER) and response time were measured. Due to the structure of the Web Speech API, it was not possible to measure response time for this implementation. The result of the benchmark shows that Web Speech API consistently outperforms the other methods in terms of WER. The response times of the browser implementation could not keep up with the other methods implemented and is not in the range of acceptable results. The recommended implementation for Knightec is to use a server-based implementation while for the general case Web Speech API is the best alternative.
37

Stenmjölets inverkan i en kall asfaltsmassa hos bearbetbarheten, brytningsprocessen och täckningsgraden / Rock Flours Impact in a Cold Asphalt Mixture of Workability, the Breaking Process and Coverage

Teikari, Michael January 2017 (has links)
I kallasfalt med hårt basbitumen bryter vattnet ut ur bitumenet och bindemedlet i massan stelnar olika snabbt beroende på vilket stenmaterial asfaltsmassan består av. Om brytningen sker innan utläggning blir massan obearbetbar, målet är att brytningsprocessen ska kunna styras genom att massans brytningsprocess skall vidta när energi tillförs i massan via en vibratorsvält som packar massan. Från tidigare tester är hypotesen att det är stenmjölet som styr brytningsprocessen och bearbetbarheten. Idag är problemet att massan bryter för snabbt och den blir svårbearbetad vid läggning och resultatet blir dåligt. I arbetet undersöks också täckningsgrad och styvhetsmodulen. Arbetet är utfört på NCC:s väglaboratorium i Umeå 4/4 - 15/4 2016. Sex stycken stenmaterial från olika bergtäkter med olika egenskaper testas. Bergtäkterna är belägna från Närke upptill Västerbotten vid namn Dylta, Råsta, Sälgsjön, Stöningsberget, Ljusberget och Supartallen.  Fraktion 0-16 mm från en och varje bergtäkt handblandas, två asfaltsprover tillverkas enligt två olika mix- och bedömningsmetoder. Täckningsgrad, bearbetningsegenskaperna och brytningprocessen bedöms och antecknas i en matris. I nästa steg ersätts ordinarie stenmjöl (0-2 mm) med stenmjöl från en bergtäkt med ett tidigare känt bra stenmaterial. Med resterande makadam (2-16 mm) görs samma process om på nytt och resultatet analyseras och jämförs. En asfaltsmassa med stenmjöl från bergtäkten Supartallen och makadam från bergtäkten Stöningsberget visar förbättrade resultat i jämförelse med fraktion 0-16 mm från bergtäkten Stöningsberget och vidare laborationstester görs på denna blandning av stenmaterial. Dessa bergtäkter ligger geografiskt nära varandra, drygt 25 mil, vilket också motiverar till att välja dessa stenmaterial som i framtiden kanske kommer användas fullskaligt. Från dessa blandningar tillverkas asfaltskroppar genom packning med gyratorisk packare och styvhetmodulen testas. I ekonomisk och miljömässig vinning görs ett försök med en restprodukt från stålindustrin. Ett försök utförs med skänkslagg (0-2 mm). Slaggen blandas med stenmaterial från bergtäkten Supartallen där bearbetningen, brytprocessen och täckningsgraden undersöks. Detta är ett inledande försök för att få en fingervisning hur asfaltsmassan beter sig blandat med skänkslagg. Utvärdering av resultaten gjordes med hjälp av en matriser där varje asfaltsblandning ingår. Asfaltsblandningen betygsätts enligt bearbetbarhet, täckningsgrad och brytningskontroll. Skålen analyserades och slutsatser om brytningen kunde dras. Genom sönderdelning av massan kunde observationer göras om hur brytningen av massan har kommit, även täckningsgraden analyseras. Resultaten pekar på att stenmjölet har en betydande roll i täckningsgrad, brytning vid blandning och brytningsprocessen, dock ingen skillnad på styvhetsmodulen. Att enbart stenmjölet ska vara problemet är inte hela sanningen men har som sagt en betydande roll. Just att stenmaterial från Supartallen är så pass bra för beaktas och använda sig av stenmaterial med liknande egenskaper och inte använda stenmaterial med för höga halter SiO2 då det ger dålig täckningsgrad. Att massan bryter för snabbt kan bero på halten bindemedel, vidare laborationer bör göras med halten bindemedel i fokus. / In cold asphalt with hard base-bitumen the water breaks out from the bitumen and the binder in the mass solidifies at different rates depending on the containing stone material in the asphalt. If the break occurs before laying the mass becomes unworkable. The intention by this examination is control this process by the mass extraction process should take place when energy is supplied into the mass through a vibrator famine. From previous tests the hypothesis is that the rock flour controls the breaking-process and the workability. Today, the problem is that the mass curing process starts too early and it becomes difficult to manage and work with and the results of the paving are bad. The thesis project also examines coverage and stiffness module, it’s performed on the NCC's road laboratory in Umeå 4/4 - 15/4 in 2016. Six stone materials from different quarries with different properties are tested. Fraction 0-16 mm from each rock quarry are handmixed, two asphalt samples produced by two different mixing- and assessment-methods. Coverage, workability and the breaking-process are assessed and recorded in a matrix. Next, replaced ordinary rock flour (0-2 mm) with rock flour from a quarry of a previously known good stone material. With the remaining makadam (2-16 mm) the process are made all over again and the results are analyzed and compared. An asphalt with rock flour from quarry Supartallen and the remaining makadam from quarry Stöningsberget shows improved results compared with fraction 0-16 mm from quarry Stöningsberget and further tests in laboratory are made with this mixture. These quarries are geographically fairly close to each other, 250 km (about 150 miles), which motivates to choose these quarries for a full scale-test in the future. From these mixtures asphalt-bodies are manufactured by gasket with gyratory compactor and stiffness module is examined. In the economic and environmental profit attempts are made with a residue from the steel industry. An attempt carried out with the slag from ladles (0-2 mm). The slag is mixed with stone material from quarry Supartallen where workability, breaking process and coverage is investigated. This is an initial attempt to introduce how the asphalt mixture behaves mixed with the slag. An evaluation was made with matrix where each asphalt mixture included. The asphalt mixture is graded according to machinability, coverage and breaking control. The bowl was analyzed and conclusions about the break could be drawn. Analyzes of the curing process and coverage was made by dismemberment of the mass. The results indicate that the rock flour has a significant role in coverage, extraction at mixing and mining process, however, the stiffness module has the same results in all mixtures. The mere stone flour shouldn’t be the problem but still has a significant role. At the aggregate of Supartallen is so good for the consideration and use of stone materials with similar properties and not rock material with a high amount of SiO2 as it provides poor coverage. The mass harden too quickly may depend on the content of binder, further laboratory work should be done with the binder content in focus.
38

Rôles physiologiques des protéines ASR à l'égard de la signalisation, du transport et du métabolisme des sucres dans deux modèles cellulaires de vigne / Physiological functions of ASR proteins regarding sugar signaling, transport and metabolism in two cell culture models in grapevine

Parrilla, Jonathan 27 March 2015 (has links)
Les sucres, sont des signaux métaboliques, impliqués dans le développement des plantes et leurs réponses aux contraintes du milieu. Les transporteurs de sucres se révèlent à la fois acteurs de la répartition des sucres et cibles de leur signalisation. L'ASR (ABA, Stress and Ripening) de la Vigne, VvMSA, étant identifiée comme protéine régulatrice de l'expression génique du transporteur d’hexoses VvHT1, l'objectif de la thèse est d'appréhender ses rôles physiologiques dans une démarche de biologie intégrée.Le premier axe a été dédié à la mise en place des modèles biologiques, des cellules embryogènes et non embryogènes de Vigne, issues du même fond génétique mais cultivées sur deux sources de carbone différentes. La caractérisation des cinétiques de prolifération et l'analyse des métabolomes ont mis en évidence leur sensibilité/tolérance différentielle à la carence en sucres. Le deuxième axe a porté sur la régulation de VvHT1 dans les deux types cellulaires sauvages et leurs mutants de surexpression/répression de VvMSA. L'approche pharmacologique utilisant des analogues du glucose, l'analyse de l'expression génique, le transport du glucose et l'activité des enzymes de la glycolyse indiquent que VvMSA affecte l'expression de VvHT1 par la voie dépendante du métabolisme du glucose. Le troisième volet a été réalisé par une approche de protéomique quantitative et comparative des protéines nucléaires des cellules embryogènes sauvages et réprimées pour VvMSA. Les protéines à expression significativement affectée par l'absence de l'ASR, laissent entrevoir un nouveau rôle à l'interconnexion des réponses métaboliques aux stress et la régulation épigénétique de l'expression génique. / Sugars are metabolic signals involved in plant development and responses to environmental cues. Sugar transporters are both actors of sugar partitioning and targets of sugar signaling. As Grape ASR (ABA, Stress, Ripening), VvMSA, is identified as a regulatory protein controlling gene expression of the hexose transporter VvHT1, the aim of the PhD thesis is to assess its physiological functions by an integrative biology approach. The first part of the study consisted in the establishment of biological models, embryogenic and non embryogenic grape cells, sharing the same genetic background but growing on distinct carbon sources. The characterization of the proliferation kinetics and metabolomes of both cell types revealed differences in their sensitivity/tolerance to sugar starvation.The second objective was focused on VvHT1 expression regulation in both cell types and their mutants overexpressing or silenced for VvMSA. The pharmacological approach using glucose analogues, coupled to the analysis of gene expression, glucose transport and glycolytic enzymes activity, suggest that VvMSA affects VvHT1 expression through a glucose metabolism dependent pathway.The third research axis was carried out through a quantitative and comparative proteomic analysis of nuclear proteins in embryogenic wild type and VvMSA silenced cells. Proteins whose expression is affected by ASR repression suggest a new functional role of VvMSA at the interplay between metabolic responses to stress and epigenetic regulation of gene expression.
39

Etude du comportement d'un alliage chromino-formeur comme matériau d'interconnecteur pour l'Electrolyse à Haute Température / Study of a chromia-forming alloy behavior as interconnect material for High Temperature Vapor Electrolysis

Guillou, Sebastien 01 December 2011 (has links)
Dans les systèmes d’Electrolyse Haute Température (EHT), le matériau choisi comme interconnecteur doit avoir une bonne résistance à la corrosion sous air et sous mélange H2/H2O à 800 °C, et maintenir une bonne conductivité sur de longues durées. Dans ce cadre, l’objectif de ce travail était, d’une part, d’évaluer un alliage ferritique commercial (l’alliage K41X) comme matériau d’interconnecteur pour l’application EHT. Dans ce but, ont été mis en place des essais d’oxydation en four et en thermoblance pour accéder aux cinétiques d’oxydation, et des mesures de résistivité pour évaluer le paramètre ASR (Area Specific Resistance) à 800°C. D’autre part, l’étude a permis d’apporter des éléments de compréhension plus fondamentaux sur les mécanismes d’oxydation des alliages chromino-formeurs, en particulier sous mélange H2-H2O, par le biais d’essais et de caractérisations spécifiques (Photoélectrochimie, traçage isotopique, essais de longues durées). Cette double stratégie est également appliquée pour l’étude d’une solution de revêtement (obtenu à l’aide de la MOCVD) basée sur l’oxyde pérovskite LaCrO3 qui présente des propriétés de conductivité élevée particulièrement intéressante en vue de l’application EHT. Ainsi, cette étude amène également des éléments de compréhension sur le rôle du lanthane comme élément réactif dont l’effet est souvent discuté dans la littérature. Pour les deux milieux, à 800°C, la couche d’oxyde formée est une couche duplexe Cr2O3/(Mn,Cr)3O4 , recouverte dans le cas du mélange H2-H2O par une fine couche d’oxyde spinelle Mn2TiO4 . Sous air, le mécanisme de croissance déterminé ici est cationique, en accord avec la littérature. La présence d’un revêtement LaCrO3 ne modifie pas ce mécanisme mais ralentit la cinétique de croissance de la couche sur les premières centaines d’heure. De plus, le revêtement améliore l’adhérence et la conductivité de la couche d’oxyde. Sous mélange H2-H2O, le mécanisme de croissance se révèle anionique. La présence de revêtement ralentit la cinétique d’oxydation. Bien que .d’épaisseurs similaires, les couches d’oxyde présentent sous air une résistivité d’un ordre de grandeur inférieure à celle mesurée sous H2-H2O. Il est mis en évidence que la forte résistivité de l’alliage en milieu H2-H2O est liée à la présence de protons issus de la vapeur d’eau présents dans la couche d’oxyde. Le revêtement ne permet néanmoins pas d améliorer la conductivité sous H2-H2O. / In High Temperature Vapor Electrolysis (HTVE) system, the materials chosen for the interconnectors should have a good corrosion behaviour in air and in H2-H2O mixtures at 800°C, and keep a high electronic conductivity over long durations as well. In this context, the first goal of this study was to evaluate a commercial ferritic alloy (the K41X alloy) as interconnect for HTVE application. Oxidation tests in furnace and in microbalance have therefore been carried out in order to determine oxidation kinetics. Meanwhile, the Area Specific Resistance (ASR) was evaluated by Contact Resistance measurements performed at 800°C. The second objective was to improve our comprehension of chromia-forming alloys oxidation mechanism, in particular in H2/H2O mixtures. For that purpose, some specific tests have been conducted: tracer experiments, coupled with the characterization of the oxide scale by PEC (PhotoElectroChemistry). This approach has also been applied to the study of a LaCrO3 perovskite oxide coating on the K41X alloy. This phase is indeed of high interest for HTVE applications due to its high conductivity properties. This latter study leads to further understanding on the role of lanthanum as reactive element, which effect is still under discussion in literature.In both media at 800°C, the scale is composed of a Cr2O3/(Mn,Cr)3O4 duplex scale, covered in the case of H2-H2O mixture by a thin scale made of Mn2TiO4 spinel. In air, the growth mechanism is found to be cationic, in agreement with literature. The LaCrO3 coating does not modify the direction of scale growth but lowers the growth kinetics during the first hundreds hours. Moreover, with the coating, the scale adherence is favored and the conductivity appears to be slightly higher. In the H2-H2O mixture, the growth mechanism is found to be anionic. The LaCrO3 coating diminishes the oxidation kinetics. Although the scale thickness is about the same in both media, the ASR parameter is one order of magnitude higher in H2/H2O than in air. Specific contact resistance tests show that the higher resistivity in the H2/H2O mixture is closely linked to the presence of protons in the scale. Moreover, tracer experiments show that these protons come from the water molecule dissociation, and not from the H2 molecule. In H2/H2O, the LaCrO3 coating does not increase the conductivity
40

Improving Grapheme-based speech recognition through P2G transliteration / W.D. Basson

Basson, Willem Diederick January 2014 (has links)
Grapheme-based speech recognition systems are faster to develop, but typically do not reach the same level of performance as phoneme-based systems. Using Afrikaans speech recognition as a case study, we first analyse the reasons for the discrepancy in performance, before introducing a technique for improving the performance of standard grapheme-based systems. It is found that by handling a relatively small number of irregular words through phoneme-to-grapheme (P2G) transliteration – transforming the original orthography of irregular words to an ‘idealised’ orthography – grapheme-based accuracy can be improved. An analysis of speech recognition accuracy based on word categories shows that P2G transliteration succeeds in improving certain word categories in which grapheme-based systems typically perform poorly, and that the problematic categories can be identified prior to system development. An evaluation is offered of when category-based P2G transliteration is beneficial and methods to implement the technique in practice are discussed. Comparative results are obtained for a second language (Vietnamese) in order to determine whether the technique can be generalised. / MSc (Computer Science) North-West University, Vaal Triangle Campus, 2014

Page generated in 0.0509 seconds