We describe the use of text data scraped from the web to augment language models for Automatic Speech Recognition and Keyword Search for Low Resource Languages. We scrape text from multiple genres including blogs, online news, translated TED talks, and subtitles. Using linearly interpolated language models, we find that blogs and movie subtitles are more relevant for language modeling of conver...

This thesis describes 80−94GHz and 70−77GHz interpolating phase shifters and the corresponding transmitter and receiver ICs, fabricated in 65-nm CMOS and SiGe BiCMOS technologies, respectively. Lumped inductors and transformers are employed to realize small-form factor 90◦ hybrids as needed in high density phased arrays. The CMOS transmitter exhibits absolute phase and amplitude errors of 4◦ an...

Obviously navigation is one of the most complicated issues in mobile robots. Intelligent algorithms are often used for error handling in robot navigation. This Paper deals with the problem of Inertial Measurement Unit (IMU) error handling by using Extended Kalman Filter (EKF) as an Expert Algorithms. Our focus is put on the field of mobile robot navigation in the 2D environments. The main chall...

The assessment of musculo-articular stiffness (MAS) with the free-oscillation technique is a popular method with a variety of applications. This study examined the sources of variability (load applied and frequency of oscillation) when MAS is assessed. Over two testing occasions, 14 healthy men (27.7±5.2 yr, 1.82±0.04 m, 79.5±8.4 kg) were measured for isometric maximum voluntary contraction and...

in this research, the spatial distribution of fe, cu, zn and mn on agricultural lands of golestan province were evaluated using different interpolation methods such as, kriging, inverse distance weighted, local polynomial, inverse multiquadric and radial basis function. thus, 505 soil samples were provided from fields during 2008 and micronutrients rates were measured for each sample. the perfo...

This paper presents two simple formulas for approximation of the standard normal right tail probabilities, and indicates the method of computing approximations of higher order. One of the approximations relies on two simple numerical constants, has absolute error of 0.00071 and relative error 0.023; the other uses four numerical constants, has absolute error of 0.000019, the relative error is ....

Data from the U.S. Geological Survey (USGS) collocated-sampler program for the National Atmospheric Deposition Program/National Trends Network (NADP/NTN) are used to estimate the overall error of NADP/NTN measurements. Absolute errors are estimated by comparison of paired measurements from collocated instruments. Spatial and temporal differences in absolute error were identified and are consist...

We develop extremely tight novel approximations, lower bounds and upper for the Gaussian $Q$ -function offer multiple alternatives coefficient sets thereof, which are optimized in terms of four most relevant criteria: minimax absolute...