In this article, the research presents a general overview of deep learning-based AVSS (audio-visual source separation) systems. AVSS has achieved exceptional results in a number of areas, including decreasing noise levels, boosting speech recognition, and improving audio quality. The advantages and disadvantages of each deep learning model are discussed throughout the research as it reviews various current experiments on AVSS. The TCD TIMIT dataset (which contains top-notch audio and video recordings created especially for speech recognition tasks) and the Voxceleb dataset (a sizable collection of brief audio-visual clips with human speech) are just a couple of the useful datasets summarized in the paper that can be used to test A

            Semi-parametric regression models have been studied in a variety of applications and scientific fields due to their high flexibility in dealing with data that has problems, as they are characterized by the ease of interpretation of the parameter part while retaining the flexibility of the non-parametric part. The response variable or explanatory variables can have outliers, and the OLS approach have the sensitivity to outliers. To address this issue, robust (resistance) methods were used, which are less sensitive in the presence of outlier values in the data. This study aims to estimate the partial regression model using the robust estimation method with the wavel

         Polymer blended electrolytes of various concentrations of undoped PAN/PMMA (80/20, 75/25, 70/30, 65/35 and 60/40 wt%) and doped with lithium salts (LiCl, Li₂SO₄H₂O, LiNO₃, Li₂CO₃) at 20% wt have been prepared by the solution casting method using dimethylformamide as a solvent. The electrical conductivity has been carried out using an LCR meter. The results showed that the highest ionic conductivity was 2.80x10^-7 (Ω.cm)^-1 and 1.05x10^-1 (Ω.cm)^-1 at 100 kHz frequency at room temperature for undoped (60% PAN + 40% PMMA) and (80% PAN + 20% PMMA) doped with 20%wt Li₂CO₃ composite blends, respect

he assignment model represents a mathematical model that aims at expressing an important problem facing enterprises and companies in the public and private sectors, which are characterized by ensuring their activities, in order to take the appropriate decision to get the best allocation of tasks for machines or jobs or workers on the machines that he owns in order to increase profits or reduce costs and time As this model is called multi-objective assignment because it takes into account the factors of time and cost together and hence we have two goals for the assignment problem, so it is not possible to solve by the usual methods and has been resorted to the use of multiple programming The objectives were to solve the problem of

Nanofluids, liquid suspensions of nanoparticles (Np), are an effective agent to alter the wettability of oil-wet reservoirs to water-wet thus promoting hydrocarbon recovery. It can also have an application to more efficient carbon storage. We present a series of contact angle (θ) investigations on initially oil-wet calcite surfaces to quantify the performance of hydrophilic silica nanoparticles for wettability alteration. These tests are conducted at typical in-situ high pressure (CO2), temperature and salinity conditions. A high pressure–temperature (P/T) optical cell with a regulated tilted surface was used to measure the advancing and receding contact angles at the desired conditions. The results showed that silica nanofluids can alte

The massive growth of the automotive industry and the development of vehicles use lead to produce a huge amount of waste tire rubber. Rubber tires are non-biodegradable, resulting in environmental problems such as fire risks. In this search, the flexural behavior of steel fiber reinforced self-compacting concrete (SFRSCC) beams containing different percentages and sizes of waste tire rubbers were studied and compared them with the flexural behavior of SCC and SFRSCC. Micro steel fiber (straight type) with aspect ratio 65 was used in mixes. The replacement of coarse and fine aggregate was 20% and 10% with chip and crumb rubber. Also, the replacement of limestone dust and silica fume was 50%, 25%, and 12% with ground rubbe

This study investigates the digestion of cow dung (CD) for biogas production at laboratory scales. The study was carried out through anaerobic fermentation using cow dung as substrate. The digester was operated at ambient temperatures of 39.5 °C for a period of 10 days. The effect of iron powder in controlling the production of hydrogen sulfide (H2S) has been tested. The optimum concentration of iron powder was 4g/L with the highest biogas production. A Q – swatch Nd:YAG laser has been used to mix and homogenize the components of one of the six digesters and accelerate digestion. At the end of digestion, all digestions effluent was subjected to 5 laser pulses with 250mJ/pules to dispose waste biomass.

Epoxy resin has many chemical features and mechanical properties, but it has a small elongation at break, low impact strength and crack propagation resistance, i.e. it exhibits a brittle behavior. In the current study, the influence of adding kaolin with variable particle size on the mechanical properties (flexural modulus E, toughness Gc, fracture toughness Kc, hardness HB, and Wear rate WR) of epoxy resin was evaluated. Composites of epoxy with varying concentrations (0, 10, 20, 30, 40 weights %) of kaolin were prepared by hand-out method. The composites showed improved (E, Gc, Kc, HB, and WR) properties with the addition of filler. Also, similar results were observed with the decrease in particle size. In addition, in this study, mult

The inverse kinematics of redundant manipulators has infinite solutions by using conventional methods, so that, this work presents applicability of intelligent tool (artificial neural network ANN) for finding one desired solution from these solutions. The inverse analysis and trajectory planning of a three link redundant planar robot have been studied in this work using a proposed dual neural networks model (DNNM), which shows a predictable time decreasing in the training session. The effect of the number of the training sets on the DNNM output and the number of NN layers have been studied. Several trajectories have been implemented using point to point trajectory planning algorithm with DNNM and the result shows good accuracy of the end