Bojan Matovski | Ss. Cyril & Methodius University in Skopje (original) (raw)

Bojan  Matovski

Uploads

Papers by Bojan Matovski

Research paper thumbnail of A Dynamic Time Warping Based Macedonian Automatic Speech Recognition System for Smart Home Applications

The decreasing cost per processing power in commercial off-the-shelf components and the decreasin... more The decreasing cost per processing power in commercial off-the-shelf components and the decreasing size of the processing units enables Automatic Speech Recognition to be a trending topic in Embedded Systems, Internet of Things and Smart Home Applications. One of the first and most intuitive algorithms used for recognizing spoken words is Dynamic Time Warping (DTW) and some state-of-the-art speech recognition algorithms still use it during the preprocessing phase in order to increase their accuracy. This paper investigates the performance of a standalone DTW-based system for the task of recognizing isolated words using a small set of known references designed for a simple Smart Home Application. The advantages and disadvantages of this approach are also analyzed, in addition to alternative implementations of the proposed algorithm and other algorithms in general which are reviewed. Finally, the results of using a straightforward DTW approach in a speech recognition system are evaluated.

Drafts by Bojan Matovski

Research paper thumbnail of IMAGE DEBLURRING AND SUPERRESOLUTION FOR SINGLE PARTICLE ANALYSIS

This work presents an approach that utilizes well established methods for image restoration using... more This work presents an approach that utilizes well established methods for image restoration using both deconvolution and super-resolution algorithms in single particle analysis performed in electron microscopy. In order to achieve satisfactory results, there is a necessity to estimate the point-spread function that simulates the negative effect of the electronic microscope combined with the 3D-reconstruction process, both simulated with the XMIPP software. Therefore, prior to the process of restoration, blurred and noisy molecule structures were created from original molecule model by using the already mentioned software. Reliable point-spread function is found by implementing inverse Fourier transform between the original and appropriate molecule models. For the step when deconvolution is performed, the blind deconvolution algorithm implemented in Matlab is utilized. In the final step, in order to increase the Signal-to-noise ratio (SNR) and to achieve more details, super-resolution is performed by fusing the information of five realizations from a given particle.

Research paper thumbnail of A Dynamic Time Warping Based Macedonian Automatic Speech Recognition System for Smart Home Applications

The decreasing cost per processing power in commercial off-the-shelf components and the decreasin... more The decreasing cost per processing power in commercial off-the-shelf components and the decreasing size of the processing units enables Automatic Speech Recognition to be a trending topic in Embedded Systems, Internet of Things and Smart Home Applications. One of the first and most intuitive algorithms used for recognizing spoken words is Dynamic Time Warping (DTW) and some state-of-the-art speech recognition algorithms still use it during the preprocessing phase in order to increase their accuracy. This paper investigates the performance of a standalone DTW-based system for the task of recognizing isolated words using a small set of known references designed for a simple Smart Home Application. The advantages and disadvantages of this approach are also analyzed, in addition to alternative implementations of the proposed algorithm and other algorithms in general which are reviewed. Finally, the results of using a straightforward DTW approach in a speech recognition system are evaluated.

Research paper thumbnail of IMAGE DEBLURRING AND SUPERRESOLUTION FOR SINGLE PARTICLE ANALYSIS

This work presents an approach that utilizes well established methods for image restoration using... more This work presents an approach that utilizes well established methods for image restoration using both deconvolution and super-resolution algorithms in single particle analysis performed in electron microscopy. In order to achieve satisfactory results, there is a necessity to estimate the point-spread function that simulates the negative effect of the electronic microscope combined with the 3D-reconstruction process, both simulated with the XMIPP software. Therefore, prior to the process of restoration, blurred and noisy molecule structures were created from original molecule model by using the already mentioned software. Reliable point-spread function is found by implementing inverse Fourier transform between the original and appropriate molecule models. For the step when deconvolution is performed, the blind deconvolution algorithm implemented in Matlab is utilized. In the final step, in order to increase the Signal-to-noise ratio (SNR) and to achieve more details, super-resolution is performed by fusing the information of five realizations from a given particle.

Log In