Browsing by Author "Ozaydin, S."

Now showing 1 - 7 of 7

Citation - Scopus: 0
A Graphical Speech Analysis Teaching Tool
(Institute of Electrical and Electronics Engineers Inc., 2022) Ozaydin, S.; Özaydın, Selma; 253019; Çankaya Meslek Yüksekokulu
The widespread use of digital speech processing in today's technologies causes many electronics and computer engineering students to need a basic background in these subjects. The paper describes a toolbox designed to support undergraduate or graduate level courses on speech processing. The proposed educational toolbox is designed as a virtual lab for basic operations in digital speech processing-based courses. This graphical user interface (GUI) based speech analysis algorithm is built with six main function modules, which are signal input, noise addition, up-sampling/down-sampling, time domain feature analysis, pitch detection and frequency domain analysis. The toolbox involves different operations for measuring important speech feature parameters such as pitch, energy, zero-crossing ratio, FFT and power spectrum of an input speech signal. The toolbox has also been developed to easily manipulate and add some other possible speech processing methods. It is thought that the tool will make it easier for students to understand the methods that form the basis of digital speech processing, increase the interest in the lesson with its visual outputs, and allow new methods to be added easily when desired thanks to its simple and modular structure. The main aim of this paper to show how such a tool facilitates students understanding of technical concepts introduced in speech courses. © 2022 IEEE.
Citation - WoS: 0
Citation - Scopus: 1
An isolated word speaker recognition system
(Institute of Electrical and Electronics Engineers Inc., 2017) Ozaydin, S.; Özaydın, Selma; Ozaydin, S.; 253019; Çankaya Meslek Yüksekokulu
The paper presents a design of an isolated word speaker recognizer system by using the Hidden Markov Model. HTK toolkit is used as a design tool. The system is operated on a closed set of speakers and with a limited vocabulary of words. Digit utterances from one to ten with ten speakers (7 male, 3 female) are used as dataset in the system. It consists of isolated words that are separated by silences. Each speaker reads each word ten times. Six of them are used in training and the remaining are used in the test phase. The Mel cepstral coefficients are used in order to design the acoustic features and HMM models are constructed. A threshold calculation is performed for each speaker. After threshold adjustment, tests are performed for the performance evaluation and accuracy rates are calculated for each user. The system resulted in changing levels of recognition accuracy for each speaker and each word. © 2017 IEEE.
Citation - Scopus: 0
Design and Comparison of Vector Quantization Codebooks for Narrowband Speech Coding
(Horizon Research Publishing, 2019) Faraj, H.; Özaydın, Selma; Ozaydin, S.; 253019; Çankaya Meslek Yüksekokulu
Vector quantization codebook algorithms are used for coding of narrow band speech signals. Multi-stage vector quantization and split vector quantization methods are two important techniques used for coding of narrowband speech signals and these methods are very popular due to the high bit rate minimization during coding of the signals. This paper presents performance measurements of multistage vector quantization and split vector quantization methods. We used line spectral frequencies for coding of the speech signals in codebook tables so as to ensure filter stability after quantization. The codebooks were generated by using the Linde-Buzo-Gray (LBG) algorithm. The tests were performed by selecting large amount of input data in training and test stages and to evaluate noise robustness of the methods, both noisy and clean speech signals were used. As a result, different codebooks were designed and tested in many stages and different bit rates to measure quantization performance. It is measured in terms of spectral distortion evaluation. We obtained the best result in 24bit multistage vector quantization codebook with a spectral distortion less than 1 dB for clean speech training data input. When we compared multistage and split vector quantization codebook spectral distortion results, multistage codebooks gave better performance in each option. © 2019 Horizon Research Publishing.
Citation - Scopus: 1
Design of a Voice Activity Detection Algorithm based on Logarithmic Signal Energy
(Institute of Electrical and Electronics Engineers Inc., 2022) Ozaydin, S.; Özaydın, Selma; 253019; Çankaya Meslek Yüksekokulu
This article presents a new method for calculating the signal energies of speech segments in voice activity detection algorithms. In the study, the μ-law signal compression method is adapted to calculate short-term signal energies. A simple voice activity detection (VAD) algorithm is designed to demonstrate the effectiveness of the proposed method. The same VAD algorithm was also run with two different conventional energy calculation formulas and the performance of each VAD was evaluated using time-domain short-time energy features. The G729 standard VAD algorithm was also used for performance comparison. During the test of the analyzed detectors, many kinds of input speech signals with various types of background environmental noise, such as restaurants, vehicles, and streets, were tested. Using the new energy calculation method, the VAD detector has improved detection accuracy compared to VAD detectors based on the other two energy methods and was able to effectively identify voice-active regions even in noisy conditions at low SNR levels. The results revealed that the VAD detector designed with the proposed new energy calculation formula outperforms traditional energy-based voice activity detection methods and provides noticeable increases in detection rate even under adverse conditions. © 2022 IEEE.
Citation - Scopus: 0
ECG Signal Denoising with SciLab
(Ismail Saritas, 2023) Ahmad, I.; Özaydın, Selma; Ozaydin, S.; 253019; Çankaya Meslek Yüksekokulu
This paper presents a study on de-noising electrocardiogram (ECG) signals using Scilab, an open-source software package known for its signal processing capabilities. ECG signals are often contaminated by various noise sources, which can reduce the accurate diagnosis and monitoring of heart health. In this work, digital signal processing methods such as Finite Impulse Response (FIR) and Infinite Impulse Response (IIR) filters are used to effectively suppress noise while preserving the essential features of the ECG waveform. We explore main noise sources that commonly affect ECG recordings, such as baseline wandering noise, power-line interference, and muscle artifacts, and discuss their respective challenges. The de-noising methods has been extensively evaluated and demonstrated its ability to improve signal quality and diagnostic accuracy by eliminating noise artifacts. The results highlight Scilab's potential for de-noising ECG signals and its importance in improving patient care and biomedical signal processing applications. The efficacy of the de-noising methods is thoroughly evaluated through comparative analyses with other commonly used de-noising approaches. Experimental results demonstrate its superiority in preserving the QRS complex while efficiently eliminating noise artifacts, leading to more accurate and reliable diagnostic information. In conclusion, this paper presents a comprehensive study on de-noising ECG signals using Scilab, offering a valuable contribution to the field of biomedical signal processing. Researchers and practitioners in the domain of ECG signal processing can benefit from the insights and techniques presented herein to advance their studies and further applications. © 2023, Ismail Saritas. All rights reserved.
Citation - Scopus: 0
A Harmonic-Based Musical Scaling Method With Natural Number Frequencies
(Genc Bilge Yayincilik, 2025) Ozaydin, S.
General acceptance arises from the most convincing method among the available options. Similarly, while the Western chromatic scale is the most widely used system today, it has limitations in representing harmonious intervals, microtonal performances, and the weak resonant effects of fractional frequencies This study introduces the Safir method, a novel approach to redefining musical note frequencies within an octave interval. Unlike traditional scales, Safir employs natural number-based values, ensuring more harmonious intervals and enhanced tuning consistency. A key strength of Safir lies in its ability to overcome the limitations of conventional tuning systems. The Safir method enhances spectral coherence by aligning note frequencies with the harmonic distribution of the Fourier series and strengthening the resonance effect through natural frequencies. This method has significant potential for various applications including music, speech and signal processing, spectral leakage reduction, and healthcare. Four key advantages of the Safir scale system are its its alignment with the harmonic series,, the strong resonant effect of note frequencies derived from natural numbers, the suppression of dissonant intervals in higher frequencies across the octave band, and its linear spacing within the octave, which ensures minimal deviation from compatible intervals even in microtonal divisions. This novel method represents a major advancement in tuning and musical scales. By providing a more precise, harmonious, and resonant frequency system, Safir addresses key shortcomings of traditional musical scales and opens new possibilities in both theoretical and practical domains. © 2025, Genc Bilge Yayincilik. All rights reserved.
Citation - Scopus: 5
Speech Denoising with Maximal Overlap Discrete Wavelet Transform
(Institute of Electrical and Electronics Engineers Inc., 2022) Alak, I.K.; Özaydın, Selma; Ozaydin, S.; 253019; Çankaya Meslek Yüksekokulu
In this paper, the effectiveness of the maximum overlapping discrete wavelet transform (MODWT) method on denoising the speech signal is tested and examined. Ensuring the intelligibility of the speech signal in noisy environments by separating it from the noise is a widely researched topic today. On the other hand, being able to recover the original speech from the noisy signal with minimal distortion is a challenge due to the difficulties in removing the background noise. Numerous factors in environmental noise environments can interfere with the signal. In this study, the performance of some discrete wavelets transform methods is experimentally analyzed using different wavelet filters. The analysis program was carried out in the MATLAB environment. As the input noise speech signal, speech sounds containing different environmental background noises (train, car, station, plane, etc.) were analyzed. During the tests, these noisy input signals were filtered out from the speech signal by wavelet analysis. The input noisy speech signal is decomposed into wavelet coefficients with different thresholding methods. The reconstructed speech was compared by measuring the signal-to-noise ratio (SNR) values between the noisy input signal and the smoothed output signals. The scientific contributions of the study include a detailed comparative analysis of the performances of various wavelet methods against different background environmental noises. © 2022 IEEE.