Gökay Dişken
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Gökay Dişken has a total of 15 co-authors. Here are the top co-authors by number of shared publications:
- Zekeriya Tufekci
Çukurova University
6 shared publications - Ulus Çevik
Department of Electrical and Electronics Engineering, Cukurova University, Adana, Turkey
3 shared publications - Baris Msc Aydin
Department of Electrical and Electronics Engineering, Adana Alparslan Turkes Science and Technology University, Adana, Turkey
1 shared publication - Cihat Ediz Akbaba
Adana Alparslan Türkeş Science and Technology University
1 shared publication - Davood Asadi
School of Engineering, University of the West of England, Frenchay Campus, Bristol, UK
1 shared publication - Evren Ekmekci
Department of Electronics and Communication Engineering, Suleyman Demirel University, Isparta, Turkey
1 shared publication
Latest Publications
- Open Access
- 11 July 2026
- Published by Elsevier in Industrial Crops and Products
- Vol. 250
Explainable transformer with optimized attention for industrial oilseed groundnut leaf stress recognition
- 11 July 2026
- Published by Elsevier in Industrial Crops and Products
- Vol. 250
Open AccessExplainable transformer with optimized attention for industrial oilseed groundnut leaf stress recognition
- 8 October 2025
- Published by Elsevier in Computers and Electrical Engineering
- Vol. 128
Delta feature maps with application to spoofed speech detection
- 8 October 2025
- Published by Elsevier in Computers and Electrical Engineering
- Vol. 128
Delta feature maps with application to spoofed speech detection
- Open Access
- 2 July 2025
- Published by Cukurova Universitesi Muhendislik-Mimarlik Fakultesi Dergisi in Çukurova Üniversitesi Mühendislik Fakültesi Dergisi
- Vol. 40 (2) , 375-386
An Investigation on Design Criteria of Heat Exchangers by Using Tree Models of Machine Learning Methods
Heat exchangers are critical components widely used in various industries such as chemical processing, automotive, and HVAC. The evaluation and optimization of heat exchanger design criteria play a vital role in improving industrial applications. Tree-based machine learning models offer a powerful alternative to time-consuming numerical solutions by enabling optimization and classification predictions for problems involving small, medium, or large datasets. This study aims to analyze heat exchanger design criteria using tree-based machine learning models and to identify the most suitable model for each design parameter. As a result, it has been evaluated that the XGBoost model provides effective solutions for design criteria such as heat transfer rate, safety, and reliability; the AdaBoost model is more suitable for criteria such as exchanger type and ease of maintenance; and the RF model performs well for cost and pumping power. It is anticipated that in the future, analyzing heat exchanger design parameters using various machine learning approaches will enable the development of more cost-effective and efficient heat exchangers.
- 2 July 2025
- Published by Cukurova Universitesi Muhendislik-Mimarlik Fakultesi Dergisi in Çukurova Üniversitesi Mühendislik Fakültesi Dergisi
- Vol. 40 (2) , 375-386
Open AccessAn Investigation on Design Criteria of Heat Exchangers by Using Tree Models of Machine Learning Methods
Heat exchangers are critical components widely used in various industries such as chemical processing, automotive, and HVAC. The evaluation and optimization of heat exchanger design criteria play a vital role in improving industrial applications. Tree-based machine learning models offer a powerful alternative to time-consuming numerical solutions by enabling optimization and classification predictions for problems involving small, medium, or large datasets. This study aims to analyze heat exchanger design criteria using tree-based machine learning models and to identify the most suitable model for each design parameter. As a result, it has been evaluated that the XGBoost model provides effective solutions for design criteria such as heat transfer rate, safety, and reliability; the AdaBoost model is more suitable for criteria such as exchanger type and ease of maintenance; and the RF model performs well for cost and pumping power. It is anticipated that in the future, analyzing heat exchanger design parameters using various machine learning approaches will enable the development of more cost-effective and efficient heat exchangers.
- Open Access
- 26 July 2024
- Published by The Scientific and Technological Research Council of Turkey (TUBITAK-ULAKBIM) - DIGITAL COMMONS JOURNALS in Turkish Journal of Electrical Engineering and Computer Sciences
- Vol. 32 (4) , 590-604
Multi-label voice disorder classification using raw waveforms
Automated voice disorder systems that distinguish pathological voices from healthy ones have been developed with the aid of machine learning methods. Both clinicians and patients can benefit from these systems as they provide many advantages, compared to the invasive techniques. These systems can produce binary (healthy/pathological) or multi-class (healthy/selected pathologies) decisions. However, multiple disorders might exist in an individual’s voice. Multi-label classification should be considered in such cases. By this time, only a single report is available on this topic, where hand-crafted features were used, and a data augmentation technique was utilized to overcome class imbalances. In this study, a similar experimental setup is followed to investigate the suitability of raw voice signals as inputs for multi-label classification. A deep learning model which consists of residual blocks and a novel gating mechanism is proposed. The gating mechanism weighs the channels of a residual block’s output based on both its output and the previous layer’s output. Using a SincNet filterbank that operates directly on the raw waveform as the initial layer, 0.99 accuracy and 0.98 F1 score were observed for natural /a/ vowels of Saarbruecken Voice Database with time domain augmentation to balance the class samples. On the other hand, reducing the number of augmented samples decreased the performance for both systems, indicating the need for a balanced dataset to avoid oversampling underrepresented classes. The proposed architecture performed consistently better than ResNet18 with deep connected attention, which verified the effectiveness of the proposed gating mechanism.
- 26 July 2024
- Published by The Scientific and Technological Research Council of Turkey (TUBITAK-ULAKBIM) - DIGITAL COMMONS JOURNALS in Turkish Journal of Electrical Engineering and Computer Sciences
- Vol. 32 (4) , 590-604
Open AccessMulti-label voice disorder classification using raw waveforms
Automated voice disorder systems that distinguish pathological voices from healthy ones have been developed with the aid of machine learning methods. Both clinicians and patients can benefit from these systems as they provide many advantages, compared to the invasive techniques. These systems can produce binary (healthy/pathological) or multi-class (healthy/selected pathologies) decisions. However, multiple disorders might exist in an individual’s voice. Multi-label classification should be considered in such cases. By this time, only a single report is available on this topic, where hand-crafted features were used, and a data augmentation technique was utilized to overcome class imbalances. In this study, a similar experimental setup is followed to investigate the suitability of raw voice signals as inputs for multi-label classification. A deep learning model which consists of residual blocks and a novel gating mechanism is proposed. The gating mechanism weighs the channels of a residual block’s output based on both its output and the previous layer’s output. Using a SincNet filterbank that operates directly on the raw waveform as the initial layer, 0.99 accuracy and 0.98 F1 score were observed for natural /a/ vowels of Saarbruecken Voice Database with time domain augmentation to balance the class samples. On the other hand, reducing the number of augmented samples decreased the performance for both systems, indicating the need for a balanced dataset to avoid oversampling underrepresented classes. The proposed architecture performed consistently better than ResNet18 with deep connected attention, which verified the effectiveness of the proposed gating mechanism.
- Open Access
- 1 January 2024
- Published by Uludag University Journal of the Faculty of Engineering in Uludağ University Journal of The Faculty of Engineering
- Vol. 29 (1) , 191-204
INCREASING ROBUSTNESS OF I-VECTORS VIA MASKING: A CASE STUDY IN SYNTHETIC SPEECH DETECTION
Ensuring security in speaker recognition systems is crucial. In the past years, it has been demonstrated that spoofing attacks can fool these systems. In order to deal with this issue, spoof speech detection systems have been developed. While these systems have served with a good performance, their effectiveness tends to degrade under noise. Traditional speech enhancement methods are not efficient for improving performance, they even make it worse. In this research paper, performance of the noise mask obtained via a convolutional neural network structure for reducing the noise effects was investigated. The mask is used to suppress noisy regions of spectrograms in order to extract robust i-vectors. The proposed system is tested on the ASVspoof 2015 database with three different noise types and accomplished superior performance compared to the traditional systems. However, there is a loss of performance in noise types that are not encountered during training phase.
- 1 January 2024
- Published by Uludag University Journal of the Faculty of Engineering in Uludağ University Journal of The Faculty of Engineering
- Vol. 29 (1) , 191-204
Open AccessINCREASING ROBUSTNESS OF I-VECTORS VIA MASKING: A CASE STUDY IN SYNTHETIC SPEECH DETECTION
Ensuring security in speaker recognition systems is crucial. In the past years, it has been demonstrated that spoofing attacks can fool these systems. In order to deal with this issue, spoof speech detection systems have been developed. While these systems have served with a good performance, their effectiveness tends to degrade under noise. Traditional speech enhancement methods are not efficient for improving performance, they even make it worse. In this research paper, performance of the noise mask obtained via a convolutional neural network structure for reducing the noise effects was investigated. The mask is used to suppress noisy regions of spectrograms in order to extract robust i-vectors. The proposed system is tested on the ASVspoof 2015 database with three different noise types and accomplished superior performance compared to the traditional systems. However, there is a loss of performance in noise types that are not encountered during training phase.
- 31 December 2023
- Published by Plusbase Akademi Organizasyon ve Danismanlik Ltd.Sti. in International Journal of Applied Methods in Electronics and Computers
Feedforward Neural Network-Based Indoor Air Quality Detection System
Indoor air quality is crucial for the sustainability of human life quality. Therefore, improving indoor air quality is critical for enhancing life quality. In this study, an artificial intelligence-based indoor air quality monitoring system is designed. The system consists of two main parts, hardware and software. The hardware part includes a control card and various sensors. The software part includes a C-based IDE software and a feedforward network, a deep learning algorithm, for establishing the connection between the control card and the sensors. The temperature, humidity, and gas concentration values obtained from the sensors at certain intervals were fed to the feedforward network's input layer through the control card. The feedforward network consists of the input layer, hidden layer, and output layer, and the decision on whether the air quality is normal or not was made at the output. The system described in this study is intended to provide a real-time monitoring solution for indoor air quality. By using a feedforward neural network, the system is able to learn patterns in the sensor data and detect changes in air quality that may indicate a problem. The system can be customized to suit the needs of different environments and can be used in a variety of settings, including homes, offices, and public spaces. Ultimately, the goal of the system is to improve human health and well-being by ensuring that indoor air quality is at a safe and healthy level.
- 31 December 2023
- Published by Plusbase Akademi Organizasyon ve Danismanlik Ltd.Sti. in International Journal of Applied Methods in Electronics and Computers
Feedforward Neural Network-Based Indoor Air Quality Detection System
Indoor air quality is crucial for the sustainability of human life quality. Therefore, improving indoor air quality is critical for enhancing life quality. In this study, an artificial intelligence-based indoor air quality monitoring system is designed. The system consists of two main parts, hardware and software. The hardware part includes a control card and various sensors. The software part includes a C-based IDE software and a feedforward network, a deep learning algorithm, for establishing the connection between the control card and the sensors. The temperature, humidity, and gas concentration values obtained from the sensors at certain intervals were fed to the feedforward network's input layer through the control card. The feedforward network consists of the input layer, hidden layer, and output layer, and the decision on whether the air quality is normal or not was made at the output. The system described in this study is intended to provide a real-time monitoring solution for indoor air quality. By using a feedforward neural network, the system is able to learn patterns in the sensor data and detect changes in air quality that may indicate a problem. The system can be customized to suit the needs of different environments and can be used in a variety of settings, including homes, offices, and public spaces. Ultimately, the goal of the system is to improve human health and well-being by ensuring that indoor air quality is at a safe and healthy level.
- 16 December 2023
- Published by Elsevier in Computer Speech & Language
- Vol. 85
Complementary regional energy features for spoofed speech detection
- 16 December 2023
- Published by Elsevier in Computer Speech & Language
- Vol. 85
Complementary regional energy features for spoofed speech detection
- Open Access
- 1 December 2023
- Published by Gazi University Journal of Science in Gazi University Journal of Science
- Vol. 36 (4) , 1553-1561
Robust Spoofed Speech Detection with Denoised I-vectors
Spoofed speech detection is recently gaining attention of the researchers as speaker verification is shown to be vulnerable to spoofing attacks such as voice conversion, speech synthesis, replay, and impersonation. Although various different methods have been proposed to detect spoofed speech, their performances decrease dramatically under the mismatched conditions due to the additive or reverberant noises. Conventional speech enhancement methods fail to recover the performance gap, hence more advanced techniques seem to be necessary to solve the noisy spoofed speech detection problem. In this work, Denoising Autoencoder (DAE) is used to obtain clean estimates of i-vectors from their noisy versions. ASVspoof 2015 database is used in the experiments with five different noise types, added to the original utterances at 0, 10, and 20 dB signal-to-noise ratios (SNR). The experimental results verified that the DAE provides a more robust spoof detection, where the conventional methods fail.
- 1 December 2023
- Published by Gazi University Journal of Science in Gazi University Journal of Science
- Vol. 36 (4) , 1553-1561
Open AccessRobust Spoofed Speech Detection with Denoised I-vectors
Spoofed speech detection is recently gaining attention of the researchers as speaker verification is shown to be vulnerable to spoofing attacks such as voice conversion, speech synthesis, replay, and impersonation. Although various different methods have been proposed to detect spoofed speech, their performances decrease dramatically under the mismatched conditions due to the additive or reverberant noises. Conventional speech enhancement methods fail to recover the performance gap, hence more advanced techniques seem to be necessary to solve the noisy spoofed speech detection problem. In this work, Denoising Autoencoder (DAE) is used to obtain clean estimates of i-vectors from their noisy versions. ASVspoof 2015 database is used in the experiments with five different noise types, added to the original utterances at 0, 10, and 20 dB signal-to-noise ratios (SNR). The experimental results verified that the DAE provides a more robust spoof detection, where the conventional methods fail.
- 4 August 2023
- Published by Elsevier in Applied Acoustics
- Vol. 211, 109568
Differential convolutional network for noise mask estimation
- 4 August 2023
- Published by Elsevier in Applied Acoustics
- Vol. 211, 109568
Differential convolutional network for noise mask estimation
- Open Access
- 29 June 2023
- Published by Celal Bayar University Journal of Science in Celal Bayar Üniversitesi Fen Bilimleri Dergisi
- Vol. 19 (2) , 167-174
Noise-Robust Spoofed Speech Detection Using Discriminative Autoencoder
Audio spoof detection gained attention of the researchers recently, as it is vital to detect spoofed speech for automatic speaker recognition systems. Publicly available datasets also accelerated the studies in this area. Many different features and classifiers have been proposed to overcome the spoofed speech detection problem, and some of them achieved considerably high performances. However, under additive noise, the spoof detection performance drops rapidly. On the other hand, number of studies about robust spoofed speech detection is very limited. The problem becomes more interesting as the conventional speech enhancement methods reportedly performed worse than no enhancement. In this work, i-vectors are used for spoof detection, and discriminative denoising autoencoder (DAE) network is used to obtain enhanced (clean) i-vectors from their noisy counterparts. Once the enhanced i-vectors are obtained, they can be treated as normal i-vectors and can be scored/classified without any modifications in the classifier part. Data from ASVspoof 2015 challenge is used with five different additive noise types, following a similar configuration of previous studies. The DAE is trained with a multicondition manner, using both clean and corrupted i-vectors. Three different noise types at various signal-to-noise ratios are used to create corrupted i-vectors, and two different noise types are used only in the test stage to simulate unknown noise conditions. Experimental results showed that the proposed DAE approach is more effective than the conventional speech enhancement methods.
- 29 June 2023
- Published by Celal Bayar University Journal of Science in Celal Bayar Üniversitesi Fen Bilimleri Dergisi
- Vol. 19 (2) , 167-174
Open AccessNoise-Robust Spoofed Speech Detection Using Discriminative Autoencoder
Audio spoof detection gained attention of the researchers recently, as it is vital to detect spoofed speech for automatic speaker recognition systems. Publicly available datasets also accelerated the studies in this area. Many different features and classifiers have been proposed to overcome the spoofed speech detection problem, and some of them achieved considerably high performances. However, under additive noise, the spoof detection performance drops rapidly. On the other hand, number of studies about robust spoofed speech detection is very limited. The problem becomes more interesting as the conventional speech enhancement methods reportedly performed worse than no enhancement. In this work, i-vectors are used for spoof detection, and discriminative denoising autoencoder (DAE) network is used to obtain enhanced (clean) i-vectors from their noisy counterparts. Once the enhanced i-vectors are obtained, they can be treated as normal i-vectors and can be scored/classified without any modifications in the classifier part. Data from ASVspoof 2015 challenge is used with five different additive noise types, following a similar configuration of previous studies. The DAE is trained with a multicondition manner, using both clean and corrupted i-vectors. Three different noise types at various signal-to-noise ratios are used to create corrupted i-vectors, and two different noise types are used only in the test stage to simulate unknown noise conditions. Experimental results showed that the proposed DAE approach is more effective than the conventional speech enhancement methods.