Sharif Digital Repository

Direct speech-to-speech translation, in which all compontents are trained jointly, is advantageous over cascaded approaches, because this method employs a simple yet effective pipeline to produce the outputs with a little inference time. Direct speech-to-speech translation models suffer from data scarcity issue, because they need parallel speech data in source and target languages. In this thesis, we present a novel direct speech-to-speech translation model to translate Persian speech to English, which is based on discrete speech units and uses a conformer-based encoder which is pretrained and a transformer-based causal decoder which uses relative position multi-head attention to do the task... 

Predicting the price that has the least error and can provide the best and highest accuracy has been one of the most challenging issues and one of the most critical concerns among capital market activists and researchers. Therefore, a model that can solve problems and provide results with high accuracy is one of the topics of interest among researchers. In this project, using time series prediction models such as ARIMA to estimate the price, variables, and indicators related to technical analysis show the behavior of traders involved in involving psychological factors for the model. By linking all of these variables to stepwise regression, we identify the best variables influencing the... 

With growth in using the Internet and entrance of online stores and internet banks to daily people life, security has become a vital part of this new world. This causes, industrial companies started to produce security devices such as fire walls for protecting network from outside attackers and Intrusion Detection/Prevention Systems (IPS/IDS) for detection of attacks inside network. IPS/IDS causes network overload due to requiring copy of packet or flow information. With introducing Software Defined Networking (SDN) architecture, there is a hope to solve the issue with combining IPS/IDS and SDN features. Therefore, we propose two novel approaches for fast network scanning and SYN Flood... 

The pursuit of a desirable life and the enhancement of citizens' life satisfaction has long been primary objectives of politicians and urban planners. While various definitions of life satisfaction have been proposed by scholars, it can be conceptualized as a comprehensive, long-term appraisal of one's quality of life, influenced by a range of factors pertaining to their living conditions. Given that mobility is an integral aspect of contemporary societies, transportation and travel mode are widely regarded as key determinants of life satisfaction. It is noteworthy that individual priorities in selecting travel modes are shaped by personal and social characteristics, which may not always... 

The vulnerability of women in the labor market is a well-established fact, but it can be long-lasting when a shock hits them. We examine women’s labor market conditions in a developing country with a substantial set of cultural norms and provide evidence that a job destructive shock persists longer among women. Specifically, probability of remaining unemployed after 20 months is about 50% for women while it is 10% for men. These findings are exacerbated for women who have not worked for more than 5 years and probably have lost their network or are new entrants in the job market. Moreover, women in the private sector, which is a more competitive market in Iran with respect to public jobs, are... 

Natural gas (NG) has emerged as an alternative energy source in transportation sector as it provides clean combustion and hence lowers exhaust pollution. NG can be stored as an adsorbed phase in porous materials and is referred to ANG. The use of adsorbent materials for storing NG, at relatively low pressure (3.5–4 MPa), is a possibility for making ANG technology competitive with CNG. Among all adsorbents, carbon nanostructures with pore diameter (0.8-1.4 nm) has shown the highest adsorption capacity of natural gas. In the present study, cellulose has been used as the carbonaceous precursor and Activated Carbon (AC) has been prepared using hydrothermal carbonization of cellulose followed by... 

The conventional methods in concrete are usually divided to destructive and nondestructive tests. Among these, nondestructive tests are more preferable because they don’t damage the structure or at least the damage doesn’t significantly affect its performance. According to precision of estimation of concrete conditions, nondestructive tests have different levels. These tests on concrete bridge decks range from chain drag to more developed test like Ground Penetration Radar in concrete The optimum ease of use, cost and the precision of results of Ultrasonic Pulse Velocity (UPV) test compared with other nondestructive methods have made it a most widely used method. The precision of results of... 

Monolithic nanocatalysts for naphta hydrodesulfurization process have been prepared and evaluated. Commercial cordierite monolith was used as a ceramic substrate. Initially, the monoliths were pretreated with nitric acid solution in order to increase specific surface area. After that, carbon nanotubes were grown through catalytic decomposition of methane over prepared monoliths. Finally, the CNT-coated monoliths were oxidized with ozone and an aqueous solution of H2O2. For the preparation of HDS Co-Mo nanocatalysts, the metal content of aqueous solution including cobalt nitrate, ammonium heptamolybdate were then impregnated onto the prepared supports. Next, samples of prepared supports and... 

Highly active unsupported MoO3 and (Co) MoO3 catalysts were synthesized utilizing the solution combustion method where the Ammonium molybdate powder and an organic additive were used as precursors. Different organic materials including Ethylene Diamin Tetra-acetic acid (EDTA), Polyethylene Glycol 200 (PEG 200), Sorbitol and Urea were used as additive in order to investigate the effect of additive structure on morphology and particle size of products. Also various reaction parameters such as the additive/Mo molar ratio, concentration of metal ion in solution, pH of the reaction and temperature of the synthesis media were changed to study effects on product morphology and size. Outcomes were... 

Optimization of energy consumption is generally important in mobile networks and in particular in automotive networks. One of the challenges of installing roadside units in automotive case networks is to provide energy to the equipment assigned to these units. In some cases, power is easily available, but in many places, the cost of energy is very high and does not have economic justification. Therefore, optimization of energy consumption in roadside unit is of great importance. Typically, vehicles limitations in terms of energy consumption for communications are far lower than that of a roadside unit. In this dissertation, part of roadside communication to vehicles is shifted to the... 

Selective oxidation of H2S to elemental sulfur in a fixed bed reactor over seven nanocatalysts has been investigated and compared to commercial claus catalyst. Among them, Al2O3-supported sodium oxide nanocatalyst prepared with wet chemical technique and Al2O3 nanocatalyst prepared with spray pyrolysis method were the most active catalysts for selective oxidation of H2S to elemental sulfur. The remaining nanocatalysts were deactivated, mainly due to the interaction with H2S and conversion into sulfides. For synthesis of Al2O3 with spray pyrolysis method we use statistical design of experiments (box-behnken design) in order to screen the effects of significant synthesizing parameters  

The low female labor force participation rate in the Iranian labor market relative to the global average and even to countries that do not differ significantly from the socio-economic characteristics of our country is one of the important questions of the Iranian labor market. In this study, we seek to answer the question of whether the economic participation rate of married women will increase if the gender gaps in wages, job findings and job losses among women and men are eliminated in the Iranian labor market. To answer this question using household expenditure and income data and labor force data, we show that between the two categories of occupations, in terms of the share of women's... 

Over the course of past decade, air pollution has been placed among the top issues chocking capital Tehran. Stationary source of air pollution is among the most important, but, less treated facors. On the year, 2014, Tehran’s air quality company addressed a number of projects in this very field could gain further governmental supports. This announcement could be a sign, showing the importance of this research. These researches in fact, have the potentioal to increase the authority’s decision making abilities in critical situations. As the models prognosticates future with acceptble accuracy, it leads to highly efficient dicissions. The results could also be used in demographic studies as... 

Fatigue crack growth can have harmful effects on the reliability rotary machine. While an early warning to crack, can be used to extend the life of these expensive machines and increase reliability.Structural Vibration Condition Monitoring as a primary fault detection method, achieved to a lot amazing result and for this reason studies vibration behavior of cracked rotational shaft is a big issue for engineers of dynamic machine.In this research, first check the fault detection methods based on structural vibration condition monitoring, mode shape methods were extracted. Then by modeling rotational shaft in computer and creation of crack on it, fault detection quality of these methods have... 

It is important to provide an appropriate and economically method with technical and engineering considerations to stabilize excavated underground spaces. The necessity of this, especially for the construction of underground metro stations located in the cities which faces many problems, is more evident. In this research, it has been tried to introduce an optimal method according to the type of area for the construction of the underground station. In order to obtain stress and strain condition in structural elements and surface deformations due to different stages of excavating, numerical modeling in FLAC3D has been used. By examining the parameters that affect the sustainability of this... 

In the present study, the effects of increasing the value of shale gas production, either due to increased production or price increases, have been studied on economic variables such as employment, wage, property value and the number of births. In this regard, the data for each of these variables was used at the county level in the United States of America between 2000 and 2011. In order to estimate the desired effects, counties are divided into two types of boom and non-boom counties. The findings indicate that increasing natural gas production will lead to more employment and wage growth in boom counties than non-boom counties. Also, during the period of increasing production, the value of... 

Co-movement between inflation and nominal interest in response to an interest rate shock, dubbed neo-fisherianism, has been the subject of contentious debate since the advent of non-inflationary ZLB environment. We design an economic model rich with banking features to understand the economic responses to permanent and long-term interest rate shocks. We find that responses to a permanent positive interest rate shock depend on the monetary authorities’ speculation of the new long-term inflation rate, with the most probable outcome being a permanent increase in inflation. Responses to a long-term positive interest rate shock depict an increase and then decrease in inflation. We also discover... 

In this study, we aim to learn representations from microscopic cell images that effectively capture the features of drugs affecting the cells, allowing us to identify effective drugs for treating a disease. We employ two parallel learning paths using predictive and generative models. Specifically, we have achieved a predictive model on the RxRx19a dataset that, unlike previous models, is interpretable, optimized, and robust to dras- tic changes in drug properties. Additionally, we have developed the first generative model on this dataset, which not only generates high-quality images but also discovers a meaningful latent space. This latent space divides the representation into relevant...