Jayalakshmy Subramaniyan
Jayalakshmy Subramaniyan has a total of 12 co-authors. Here are the top co-authors by number of shared publications:
- Gnanou Florence Sudha
Electronics and Communication Engineering, Pondicherry Engineering College, Puducherry, India
3 shared publications - Lakshmipriya Balagourouchetty
School of Engineering, Jawaharlal Nehru University, New Delhi, India
3 shared publications - B. Lakshmipriya
School of Engineering, Jawaharlal Nehru University, New Delhi, India
2 shared publications - Elizabeth Caroline Britto
IFET College of Engineering, Electronics and communication Engineering Department, Villupuram, Tamilnadu, India
1 shared publication - Janani Dhamodaran
IFET College of Engineering, Electronics and communication Engineering Department, Villupuram, Tamilnadu, India
1 shared publication - Joseph Bamidele Awotunde
University of Ilorin
1 shared publication
Latest Publications
- 15 September 2026
- Published by SAGE Publications in Journal of Intelligent & Fuzzy Systems: Applications in Engineering and Technology
Empirical mode decomposition based CNN prediction model for speech emotion classification using Mel scale spectrograms
Studies on emotion recognition systems from speech signal are comprehensively assessed in the purview of developing Human-Computer Interface (HCI) models. Traditionally, HCI models offer significant contribution into the psychological states of individuals, serving as crucial psychic response for HCI applications. Majority of the existing works in literature concentrate on classifying emotional states with several spectral and Cepstrum features using ANN and RNN based classification for various data sets. Despite that, the results are found to be comparatively low for RAVDESS dataset. Furthermore, only limited studies focus on using 3D time frequency images for predicting Speech Emotion Recognition (SER) models. In light of these aspects, the proposed work aims at recognizing the speech emotion with seven different states using deep neural net with Empirical mode decomposition (EMD).In the initial phase, the raw speech signals are decomposed into 5 different Intrinsic Mode Functions (IMFs) and are analysed in time, frequency and energy. Subsequently, all the five oscillatory modes are transformed into linear scale and Mel scale spectrograms using short time frequency transform and auditory filter banks. These spectrograms are fed as input to three different pre-trained models namely AlexNet, GoogLeNet and ResNet-50 and evaluated for all IMFs both in linear and Mel Scale. Ablation studies are also carried out with controlled baselines viz. Mel/ Linear scale spectrograms without EMD and classical Machine Learning classifiers for fair comparison. The experimental results record an improved performance in accuracy for IMF1 + Mel scale spectrograms + ResNet-50 classifier with 76.38% and linear scale of 74.25%. Additionally, 5- fold cross validation analysis is also carried out for generalizing the model and the results are reported. In comparison to the other approaches used in literature for categorizing RAVDESS dataset, the proposed method found to yield better results.
- 15 September 2026
- Published by SAGE Publications in Journal of Intelligent & Fuzzy Systems: Applications in Engineering and Technology
Empirical mode decomposition based CNN prediction model for speech emotion classification using Mel scale spectrograms
Studies on emotion recognition systems from speech signal are comprehensively assessed in the purview of developing Human-Computer Interface (HCI) models. Traditionally, HCI models offer significant contribution into the psychological states of individuals, serving as crucial psychic response for HCI applications. Majority of the existing works in literature concentrate on classifying emotional states with several spectral and Cepstrum features using ANN and RNN based classification for various data sets. Despite that, the results are found to be comparatively low for RAVDESS dataset. Furthermore, only limited studies focus on using 3D time frequency images for predicting Speech Emotion Recognition (SER) models. In light of these aspects, the proposed work aims at recognizing the speech emotion with seven different states using deep neural net with Empirical mode decomposition (EMD).In the initial phase, the raw speech signals are decomposed into 5 different Intrinsic Mode Functions (IMFs) and are analysed in time, frequency and energy. Subsequently, all the five oscillatory modes are transformed into linear scale and Mel scale spectrograms using short time frequency transform and auditory filter banks. These spectrograms are fed as input to three different pre-trained models namely AlexNet, GoogLeNet and ResNet-50 and evaluated for all IMFs both in linear and Mel Scale. Ablation studies are also carried out with controlled baselines viz. Mel/ Linear scale spectrograms without EMD and classical Machine Learning classifiers for fair comparison. The experimental results record an improved performance in accuracy for IMF1 + Mel scale spectrograms + ResNet-50 classifier with 76.38% and linear scale of 74.25%. Additionally, 5- fold cross validation analysis is also carried out for generalizing the model and the results are reported. In comparison to the other approaches used in literature for categorizing RAVDESS dataset, the proposed method found to yield better results.
- 23 May 2026
- Published by Elsevier in Computers in Biology and Medicine
- Vol. 212, 111776
Variational mode decomposition based prediction model for cough sounds using Xception-GRU classifier
- 23 May 2026
- Published by Elsevier in Computers in Biology and Medicine
- Vol. 212, 111776
Variational mode decomposition based prediction model for cough sounds using Xception-GRU classifier
- Open Access
- 3 October 2025
- Published by Springer Nature in Annals of Data Science
- p. 1-33
COPD Scope: Smart Clinical Pathways for Respiratory Illness Diagnosis and Management
Internet of Things (IoT) is playing a vital role in healthcare by automating the real time monitoring of patients seamlessly with the help of a variety of sensors. In the current scenario of smart healthcare, IoT with the integration of machine learning and deep learning algorithms is instrumental in the management of chronic diseases like chronic obstructive pulmonary diseases (COPD). IoT when deployed for COPD, supports in the monitoring of lung functioning, uninterrupted patient monitoring with personalised treatment planning and also facilitates in the implementation of telemedicine technology. This paper presents a comprehensive insight on the diagnosis and treatment strategies of COPD from multiple modalities viz. smart sensors for respiratory monitoring, auscultations and imaging-based diagnosis. An exhaustive experimentation of various machine learning and deep learning algorithms for the diagnosis of respiratory illness from the auscultation recordings corresponding to four classes: normal, crackle, Rhinchi and wheeze has been presented as a case study. Lung sounds were analysed in different dimensions by considering the signal as one-dimensional (1D) vector and multiple time frequency representations derived from the 1D signals. Statistical, spectral and cepstral features derived from the signals were classified using decision tree, support vector machine, K- nearest neighbor and random forest classifiers. In a different spectrum, convolutional neural networks were employed to perform classification of lung sounds from 1D signals as such and spectrogram, melspectrogram and scalogram representations derived from 1D signals. From the outcomes of experimentation, it is inferred that visual representations of signal frequency spectrum in the form of melspectrogram representations play a significant role in categorizing the patterns belonging to four classes of respiratory diseases. The proposed model COPDScope upon experimenting with different optimizers, it is found that Melspectrogram representation records a better performance over the other two representations.
- 3 October 2025
- Published by Springer Nature in Annals of Data Science
- p. 1-33
Open AccessCOPD Scope: Smart Clinical Pathways for Respiratory Illness Diagnosis and Management
Internet of Things (IoT) is playing a vital role in healthcare by automating the real time monitoring of patients seamlessly with the help of a variety of sensors. In the current scenario of smart healthcare, IoT with the integration of machine learning and deep learning algorithms is instrumental in the management of chronic diseases like chronic obstructive pulmonary diseases (COPD). IoT when deployed for COPD, supports in the monitoring of lung functioning, uninterrupted patient monitoring with personalised treatment planning and also facilitates in the implementation of telemedicine technology. This paper presents a comprehensive insight on the diagnosis and treatment strategies of COPD from multiple modalities viz. smart sensors for respiratory monitoring, auscultations and imaging-based diagnosis. An exhaustive experimentation of various machine learning and deep learning algorithms for the diagnosis of respiratory illness from the auscultation recordings corresponding to four classes: normal, crackle, Rhinchi and wheeze has been presented as a case study. Lung sounds were analysed in different dimensions by considering the signal as one-dimensional (1D) vector and multiple time frequency representations derived from the 1D signals. Statistical, spectral and cepstral features derived from the signals were classified using decision tree, support vector machine, K- nearest neighbor and random forest classifiers. In a different spectrum, convolutional neural networks were employed to perform classification of lung sounds from 1D signals as such and spectrogram, melspectrogram and scalogram representations derived from 1D signals. From the outcomes of experimentation, it is inferred that visual representations of signal frequency spectrum in the form of melspectrogram representations play a significant role in categorizing the patterns belonging to four classes of respiratory diseases. The proposed model COPDScope upon experimenting with different optimizers, it is found that Melspectrogram representation records a better performance over the other two representations.
- 1 September 2024
- Published by SPIE-Intl Soc Optical Eng in Optical Engineering
- Vol. 63 (09) , 095102-16
Design of photonic crystal semiconductor optical amplifier-based all-optical multiplexer and parameter optimization using particle swarm optimization algorithm
We propose a design leveraging photonic crystal semiconductor optical amplifiers (PhC-SOA), harnessing its unique capacity to manipulate and govern light at the nanoscale. Optimization of critical parameters to enhance performance is pursued through the utilization of the particle swarm optimization (PSO) algorithm. Initially, a multiplexer was constructed using logic gates, achieving a peak output power of 1.94e−11 mW. A PSO algorithm was employed to find the optimal PhC-SOA parameters such as length of active layer, input energy, simulation steps, full width at half maximum (FWHM), carrier recovery time, linewidth enhancement factor, saturation energy, and linear gain to maximize performance. The peak output power obtained from the PSO-optimized multiplexer is 2.9e−11 mW. This paper demonstrates the performance of PhC-SOA-based all-optical multiplexers and the effectiveness of PSO in optimizing the performance of all-optical multiplexers and contributes to the advancement of optical communication systems for high-speed applications.
- 1 September 2024
- Published by SPIE-Intl Soc Optical Eng in Optical Engineering
- Vol. 63 (09) , 095102-16
Design of photonic crystal semiconductor optical amplifier-based all-optical multiplexer and parameter optimization using particle swarm optimization algorithm
We propose a design leveraging photonic crystal semiconductor optical amplifiers (PhC-SOA), harnessing its unique capacity to manipulate and govern light at the nanoscale. Optimization of critical parameters to enhance performance is pursued through the utilization of the particle swarm optimization (PSO) algorithm. Initially, a multiplexer was constructed using logic gates, achieving a peak output power of 1.94e−11 mW. A PSO algorithm was employed to find the optimal PhC-SOA parameters such as length of active layer, input energy, simulation steps, full width at half maximum (FWHM), carrier recovery time, linewidth enhancement factor, saturation energy, and linear gain to maximize performance. The peak output power obtained from the PSO-optimized multiplexer is 2.9e−11 mW. This paper demonstrates the performance of PhC-SOA-based all-optical multiplexers and the effectiveness of PSO in optimizing the performance of all-optical multiplexers and contributes to the advancement of optical communication systems for high-speed applications.
- 16 May 2024
- Published by IGI Global Scientific Publishing
- p. 242-261
Ensembled Time Series Deep Learning Framework for Stock Market Prediction
Prediction and analysis of Stock Market plays a very important and crucial role in today's economic growth. Understanding the pattern of these financial activity and predicting its transformation and development are the most challenging areas of research in financial sectors and academic sectors. This paper explores the competency of sequential learning elements viz. long short-term memory (LSTM), bidirectional long short-term memory (BiLSTM) and gated recurrent unit (GRU) with different depths and combinations in forecasting the share market trend from the Nifty50 dataset. The experimental assessment elucidates that the ensemble prediction model built using BiLSTM-LSTM and BiLSTM-GRU by the virtue of integrating the merits of BiLSTM and LSTM/GRU layers demonstrate a better performance closer to reality with least error. Request access from your librarian to read this chapter's full text.
- 16 May 2024
- Published by IGI Global Scientific Publishing
- p. 242-261
Ensembled Time Series Deep Learning Framework for Stock Market Prediction
Prediction and analysis of Stock Market plays a very important and crucial role in today's economic growth. Understanding the pattern of these financial activity and predicting its transformation and development are the most challenging areas of research in financial sectors and academic sectors. This paper explores the competency of sequential learning elements viz. long short-term memory (LSTM), bidirectional long short-term memory (BiLSTM) and gated recurrent unit (GRU) with different depths and combinations in forecasting the share market trend from the Nifty50 dataset. The experimental assessment elucidates that the ensemble prediction model built using BiLSTM-LSTM and BiLSTM-GRU by the virtue of integrating the merits of BiLSTM and LSTM/GRU layers demonstrate a better performance closer to reality with least error. Request access from your librarian to read this chapter's full text.
- 9 March 2023
- Published by Springer Nature in International Journal of Information Technology
- Vol. 15 (3) , 1699-1708
Wavelet scattering and scalogram visualization based human brain decoding using empirical wavelet transform
- 9 March 2023
- Published by Springer Nature in International Journal of Information Technology
- Vol. 15 (3) , 1699-1708
Wavelet scattering and scalogram visualization based human brain decoding using empirical wavelet transform
- 8 October 2021
- Published by Elsevier in Computers in Biology and Medicine
- Vol. 138, 104930
Conditional GAN based augmentation for predictive modeling of respiratory signals
- 8 October 2021
- Published by Elsevier in Computers in Biology and Medicine
- Vol. 138, 104930
Conditional GAN based augmentation for predictive modeling of respiratory signals
- 23 July 2021
- Published by Springer Nature in Neural Computing and Applications
- Vol. 33 (24) , 17029-17040
GTCC-based BiLSTM deep-learning framework for respiratory sound classification using empirical mode decomposition
- 23 July 2021
- Published by Springer Nature in Neural Computing and Applications
- Vol. 33 (24) , 17029-17040
GTCC-based BiLSTM deep-learning framework for respiratory sound classification using empirical mode decomposition
- 1 March 2020
- Published by Elsevier in Artificial Intelligence in Medicine
- Vol. 103, 101809
Scalogram based prediction model for respiratory disorders using optimized convolutional neural networks
- 1 March 2020
- Published by Elsevier in Artificial Intelligence in Medicine
- Vol. 103, 101809
Scalogram based prediction model for respiratory disorders using optimized convolutional neural networks