• Digital Humanities and Cultural Heritage
• 2018-2022close
• Thesis close
• Digital Humanities and Cultural Her... close

[ES] Durante la última década, los medios de comunicación han experimentado una revolución, alejándose de la televisión convencional hacia las plataformas de contenido bajo demanda. Además, esta revolución no ha cambiado solamente la manera en la que nos entretenemos, si no también la manera en la que aprendemos. En este sentido, las plataformas de contenido educativo bajo demanda también han proliferado para proporcionar recursos educativos de diversos tipos. Estas nuevas vías de distribución de contenido han llegado con nuevos requisitos para mejorar la accesibilidad, en particular las relacionadas con las dificultades de audición y las barreras lingüísticas. Aquí radica la oportunidad para el reconocimiento automático del habla (RAH) para cumplir estos requisitos, proporcionando subtitulado automático de alta calidad. Este subtitulado proporciona una base sólida para reducir esta brecha de accesibilidad, especialmente para contenido en directo o streaming. Estos sistemas de streaming deben trabajar bajo estrictas condiciones de tiempo real, proporcionando la subtitulación tan rápido como sea posible, trabajando con un contexto limitado. Sin embargo, esta limitación puede conllevar una degradación de la calidad cuando se compara con los sistemas para contenido en diferido u offline. Esta tesis propone un sistema de RAH en streaming con baja latencia, con una calidad similar a un sistema offline. Concretamente, este trabajo describe el camino seguido desde el sistema offline híbrido inicial hasta el eficiente sistema final de reconocimiento en streaming. El primer paso es la adaptación del sistema para efectuar una sola iteración de reconocimiento haciendo uso de modelos de lenguaje estado del arte basados en redes neuronales. En los sistemas basados en múltiples iteraciones estos modelos son relegados a una segunda (o posterior) iteración por su gran coste computacional. Tras adaptar el modelo de lenguaje, el modelo acústico basado en redes neuronales también tiene que adaptarse para trabajar con un contexto limitado. La integración y la adaptación de estos modelos es ampliamente descrita en esta tesis, evaluando el sistema RAH resultante, completamente adaptado para streaming, en conjuntos de datos académicos extensamente utilizados y desafiantes tareas basadas en contenidos audiovisuales reales. Como resultado, el sistema proporciona bajas tasas de error con un reducido tiempo de respuesta, comparables al sistema offline. [CA] Durant l'última dècada, els mitjans de comunicació han experimentat una revolució, allunyant-se de la televisió convencional cap a les plataformes de contingut sota demanda. A més a més, aquesta revolució no ha canviat només la manera en la que ens entretenim, si no també la manera en la que aprenem. En aquest sentit, les plataformes de contingut educatiu sota demanda també han proliferat pera proporcionar recursos educatius de diversos tipus. Aquestes noves vies de distribució de contingut han arribat amb nous requisits per a millorar l'accessibilitat, en particular les relacionades amb les dificultats d'audició i les barreres lingüístiques. Aquí radica l'oportunitat per al reconeixement automàtic de la parla (RAH) per a complir aquests requisits, proporcionant subtitulat automàtic d'alta qualitat. Aquest subtitulat proporciona una base sòlida per a reduir aquesta bretxa d'accessibilitat, especialment per a contingut en directe o streaming. Aquests sistemes han de treballar sota estrictes condicions de temps real, proporcionant la subtitulació tan ràpid com sigui possible, treballant en un context limitat. Aquesta limitació, però, pot comportar una degradació de la qualitat quan es compara amb els sistemes per a contingut en diferit o offline. Aquesta tesi proposa un sistema de RAH en streaming amb baixa latència, amb una qualitat similar a un sistema offline. Concretament, aquest treball descriu el camí seguit des del sistema offline híbrid inicial fins l'eficient sistema final de reconeixement en streaming. El primer pas és l'adaptació del sistema per a efectuar una sola iteració de reconeixement fent servir els models de llenguatge de l'estat de l'art basat en xarxes neuronals. En els sistemes basats en múltiples iteracions aquests models son relegades a una segona (o posterior) iteració pel seu gran cost computacional. Un cop el model de llenguatge s'ha adaptat, el model acústic basat en xarxes neuronals també s'ha d'adaptar per a treballar amb un context limitat. La integració i l'adaptació d'aquests models és àmpliament descrita en aquesta tesi, avaluant el sistema RAH resultant, completament adaptat per streaming, en conjunts de dades acadèmiques àmpliament utilitzades i desafiants tasques basades en continguts audiovisuals reals. Com a resultat, el sistema proporciona baixes taxes d'error amb un reduït temps de resposta, comparables al sistema offline. [EN] Over the last decade, the media have experienced a revolution, turning away from the conventional TV in favor of on-demand platforms. In addition, this media revolution not only changed the way entertainment is conceived but also how learning is conducted. Indeed, on-demand educational platforms have also proliferated and are now providing educational resources on diverse topics. These new ways to distribute content have come along with requirements to improve accessibility, particularly related to hearing difficulties and language barriers. Here is the opportunity for automatic speech recognition (ASR) to comply with these requirements by providing high-quality automatic captioning. Automatic captioning provides a sound basis for diminishing the accessibility gap, especially for live or streaming content. To this end, streaming ASR must work under strict real-time conditions, providing captions as fast as possible, and working with limited context. However, this limited context usually leads to a quality degradation as compared to the pre-recorded or offline content. This thesis is aimed at developing low-latency streaming ASR with a quality similar to offline ASR. More precisely, it describes the path followed from an initial hybrid offline system to an efficient streaming-adapted system. The first step is to perform a single recognition pass using a state-of-the-art neural network-based language model. In conventional multi-pass systems, this model is often deferred to the second or later pass due to its computational complexity. As with the language model, the neural-based acoustic model is also properly adapted to work with limited context. The adaptation and integration of these models is thoroughly described and assessed using fully-fledged streaming systems on well-known academic and challenging real-world benchmarks. In brief, it is shown that the proposed adaptation of the language and acoustic models allows the streaming-adapted system to reach the accuracy of the initial offline system with low latency.

Roseltha Wolverton Goble (1835-1919), a farm woman in Oxford County, was a devoted diarist. In her diary collection, she preserved a detailed record of her life over six decades. Her eight diaries, written between 1857 and 1919, offer valuable insights into the evolution of her diary-writing practices. While analyses of rural women have primarily focused on how rural women conveyed identity through their actions, this study concerns itself with Roseltha’s interiority, examining how her relationship to her journals informed her evolving sense of self. By adjusting her writing habits to complement the changing rhythms and responsibilities incumbent on her as a rural farm woman, Roseltha integrated diary-writing into the fabric of her life as a young, adult, and mature woman. Through this approach, Roseltha struck a delicate balance between sense and sensibility, modifying her diaries’ format, tone, and style to continually re-establish journaling as an essential practice in her life amidst changing circumstances.

The Mediterranean Sea is a large sea almost completely enclosed by Europe, Africa, and SW Asia. Today, the continuous flow of Atlantic seawater through the Strait of Gibraltar balances the net water loss due to the excess of evaporation over input from rain and rivers and keeps the Mediterranean a marine basin with the water level equal to the global ocean level. In the late Miocene, the input of Atlantic seawater was through shallow gateways in southern Spain and northern Morocco. The exhumation of the Betic and Rif Cordilleras and orbitally-controlled regional climate change contributed to the step-by-step close these marine gateways and transformed the Mediterranean into an inhospitable evaporitic sea, starting at 5.96±0.02 Ma. The environmental and hydrological crisis that followed, known as the Messinian Salinity Crisis, peaked at ~5.55 Ma, when a 1.5-2 km-thick salt layer precipitated on the sea floor, and terminated at ~5.332 Ma, when normal marine conditions were re-established, perhaps as a consequence of catastrophic flooding at Gibraltar. The transitional interval from the giant salina to Mediterranean Sea that resembles what we know today is known as the Lago-Mare phase. Gypsum and brackish water biota from the Central Paratethys (approximately coinciding with the present-day Hungary-Slovenia) are both characteristic of Lago-Mare successions. This sedimentary and paleontological evidence suggest that the Mediterranean, during the Lago-Mare phase, experienced a complex history of inter- and extrabasinal connections, base-level fluctuations and extreme environmental and biological changes. This thesis aims to reconstruct the fiercely debated Mediterranean water level changes and identify the hydrological fluxes responsible for these Lago-Mare environmental and biological turnover events. It focuses on key sections exposed on-land and recovered from deep sea cores along an E-W transect across the Mediterranean with a multi-disciplinary approach that integrates biocyclostratigraphic analyses with high-resolution Sr-isotope ratios. The new data show that the Lago-Mare phase was a far more dynamic period of the Mediterranean history than previously thought. Large areas of the Mediterranean were occupied by an anomalohaline water mass predominantly formed by low-salinity waters from the Eastern Paratethys (i.e. the present-day Black-Caspian seas) and from the larger circum-Mediterranean rivers such as the Nile, the Rhône and currently dried North-African rivers. A small and possibly episodic Atlantic flux was a minor contributor. The regular variability of the Mediterranean’s freshwater budget with the astronomic precession combined with fluctuating Atlantic inflow and negligible outflow resulted in a Mediterranean base level that oscillated dramatically, but consistently below eustatic sea level. These precession-paced fluctuations caused the low-salinity Paratethyan biota to colonize the marginal areas of the Mediterranean at times of high base level and central areas during phases of low base level, gypsum to precipitate only in the deeper basins during the regressive phases and silled regions such as the Adriatic and the North Aegean to exist for long as some of the largest endorheic lakes.

A pesquisa investiga as relações entre processos e temas na pintura de Gerhard Richter. Selecionamos alguns trabalhos e séries de obras, tendo como base a pluralidade de abordagens que o artista cria em seu escopo de operações ao longo de seus quase 60 anos de carreira. No primeiro capítulo, verificamos como Richter desenvolve tal pluralidade em suas obras, através de um debate em torno da composição construtiva, em contraponto à de caráter informal. Discutimos, ainda, um fundamento que julgamos ser essencial no trabalho do artista: retirar o lugar do discurso pelo prisma do que é ou pode vir a ser essencial na pintura. No segundo capítulo, colocamos em pauta um sistema de binarismos que o crítico de arte Robert Storr denominou Binários Dialéticos, na obra de Richter. Trata-se de um jogo complementar entre figuração e abstração pelo viés do trabalho nos sistemas capitalista e comunista, o que incide também sobre a circulação dos objetos de arte dentro de um determinado contexto político e social. Dessa maneira, investigamos também a diferença entre as funções da arte para a política e a sociedade, de um lado, e a arte com tema político-social, de outro. Já no terceiro capítulo serão abordadas algumas séries de obras nasquais Richter utilizou temas políticos, tais como: September, o ciclo de pinturas Badeer Meinhof e a série Birkenau. Construímos aqui um histórico das diferentes abordagens e operação pictóricas em referências a temas políticos na obra de Richter. Como conclusão da pesquisa, travamos um debate em torno da ética da representação. The research investigates the relationships between processes and themes in painting and their development in the work by Gerhard Richter. We selected some works and series of works based on the plurality of approaches that the artist creates in his scope of operations throughout his almost 60 years of career. In the first chapter, we verify how Richter develops such plurality in his works through a debate around constructive composition as opposed to informal composition. we discussed still a foundation that we believe to be essential in the artist’s work: removing the place of speech through the prism of what is or could become essential in painting. In the second chapter, we put on the agenda a system of binarities that the art critic Robert Storr called Dialectical Binaries in Richter’s work. It is a complementary game between figuration and abstraction through the bias of work in the capitalist and communist system, which also affects on the circulation of art objects within a given political and social context. Of that In the same way, we also investigate the difference between the functions of art for politics and society, from on the one hand, and art with a political-social theme on the other. In the third chapter, some series of works in which Richter used themes will be approached. such as: September, the Badeer Meinhof painting cycle and the Birkenau series. We build here a history of Richter’s use of different approaches and operation pictorial references to political themes. As a conclusion of the research, we held a debate on around the ethics of representation.

The potential link between climate variability, conflict, and migration is increasingly viewed as a security issue by policy makers. Climate variability and extremes raise critical challenges to agriculture and food production all over the world, and lead to diminished coping capacity, loss of livelihoods, as well as migration flows. The essays in this dissertation raise the question of resilience to shocks in pre-industrial economies using the cases of France and Savoy during the Ancien Régime. More specifically, it documents the role of institutions and migration in reducing the vulnerability to climate shocks and hence violence. It further studies both the effect of seasonal migration as a resilience strategy against adverse economic conditions, and the socio-economic consequences of large and unexpected episode of migration.

In 1903 and 1904, E.T. Whittaker produced a pair of papers that, although sometimes considered mere mathematical statements (Barrett 1993), held important implications if viewed as physical realities. His 1903 paper united electrostatic and gravitational attraction as both resulting from longitudinal waves in the medium. His 1904 paper specifically dealt with electromagnetic waves as resulting from the interference of two such longitudinal waves as scalar potential functions. These papers, if representative of physical reality, may have important implications: gravitational lensing, gravitational waves, the Aharonov-Bohm effect, the existence of a hyperspace above or behind normal space, and the elimination of gravitational and point charge singularities. Added Introduction, Added New References