Development of speech emotion recognition system using optimized convolutional neural network

Pest and disease management plays a significant role in minimizing losses to crops, particularly in citrus fruit production. Traditional methods for detecting and classifying infected citrus fruits are complex and tasking, while Convolutional Neural Networks (CNNs) offer promising solutions but still face challenges such as high computational requirements and data dependency. Therefore, this study developed an improved convolution neural network for infected citrus fruit detection and classification system using Coati Optimization Algorithm (COA). A dataset of 1,790 citrus images, containing samples of black spot, greening spot, citrus canker, and healthy fruits, was acquired from www.kaggle.com. The images underwent preprocessing involving cropping to remove unwanted elements, conversion to grayscale to simplify processing, normalization to enhance data consistency and reduce redundancy, and filtering to minimize noise. An optimized CNN model was formulated using COA to tune the hyperparameters (weight and learning rate) of CNN to produce Coati Optimization Algorithm–based Convolutional Neural Network (COA-CNN). The preprocessed images serve as input to the COA-CNN model. The COA-CNN was used for the extraction of edges, corners, texture, patterns and shapes, and classification of citrus fruits as infected or healthy. The developed system was implemented using MATLAB R(2023a). The system’s performance was evaluated using accuracy, false positive rate, sensitivity, specificity, and recognition time. A comparative analysis of CNN and COA-CNN was also carried out. The accuracy, false positive rate, sensitivity, specificity, and recognition time for CNN were 95.83%, 6.02%, 96.63%, 93.98% and 202.17 s, respectively, while the corresponding values for COA-CNN were 96.92%, 4.22%, 97.41%, 95.78% and 136.86 s. This research showed that COA-CNN performed better and is recommended for citrus disease detection and classification systems.

Artificial intelligence (AI) and machine learning (ML) are employed to make systems smarter. Today, the speech emotion recognition (SER) system evaluates the emotional state of the speaker by investigating his/her speech signal. Emotion recognition is a challenging task for a machine. In addition, making it smarter so that the emotions are efficiently recognized by AI is equally challenging. The speech signal is quite hard to examine using signal processing methods because it consists of different frequencies and features that vary according to emotions, such as anger, fear, sadness, happiness, boredom, disgust, and surprise. Even though different algorithms are being developed for the SER, the success rates are very low according to the languages, the emotions, and the databases. In this paper, we propose a new lightweight effective SER model that has a low computational complexity and a high recognition accuracy. The suggested method uses the convolutional neural network (CNN) approach to learn the deep frequency features by using a plain rectangular filter with a modified pooling strategy that have more discriminative power for the SER. The proposed CNN model was trained on the extracted frequency features from the speech data and was then tested to predict the emotions. The proposed SER model was evaluated over two benchmarks, which included the interactive emotional dyadic motion capture (IEMOCAP) and the berlin emotional speech database (EMO-DB) speech datasets, and it obtained 77.01% and 92.02% recognition results. The experimental results demonstrated that the proposed CNN-based SER system can achieve a better recognition performance than the state-of-the-art SER systems.

Designing of VehiNet Using Convolutional Neural Networks and Deep Learning Techniques

Biometrics is a technology that identifies or verifies individuals based on unique physical or behavioural traits, offering a reliable form of authentication in sectors like healthcare, law enforcement, and security. Existing gender detection systems using fingerprints face challenges due to poor image quality and complex ridge patterns, while ConvolutionalNeural Networks (CNNs), though promising, are hindered by issues like overfitting, slow convergence, and getting trapped in local minima.Other optimisation algorithms like Genetic Algorithm (GA), Particle Swarm Optimisation (PSO)and Ant Colony Optimisation(ACO)often face issues likepremature convergence,high computational costsand poor global search ability. Whale Optimisation Algorithm (WOA) was chosen for its faster convergence, simplicity, and better balance between exploration and exploitation in optimising CNN parameters.Therefore, this study developed a fingerprint-based gender detection system through the Optimisationof CNN with Whale OptimisationAlgorithm.A dataset of 2,200 gender-labelled fingerprint images (1,320 male and 880 female) was acquired from Kaggle.com. The images underwent preprocessing involving cropping, grayscale conversion, histogram equalization for enhancement, and edge detection filtering to eliminate noise. OptimisedCNN model was formulated using Whale OptimisationAlgorithm (WOA) by tuning CNN hyperparameters: number of neurons and dropout rate. The resulting WOA-CNN was employed for feature extraction (edges, texture patterns, shapes) and detection of fingerprint images. The model was implemented in MATLAB R2023a.Performance was evaluated using accuracy, sensitivity, specificity, false positive rate, precision, and recognition time, with an 80-20% training-testing split. CNN achieved 95.86% accuracy, 96.44% sensitivity, 95.00% specificity, 5.00% false positive rate, 96.66% precision, and 99.90 s recognition time. WOA-CNN achieved 97.23% accuracy, 97.58% sensitivity, 96.70% specificity, 3.30% false positive rate, 97.80% precision, and 87.40 s recognition time.This research showed WOA-CNN outperformed CNN in all metrics. It is recommended for use in biometric authentication, security checkpoints, and forensic investigations.

An ensemble 1D-CNN-LSTM-GRU model with data augmentation for speech emotion recognition

Identification of crops is an important topic in the agricultural domain. Hyperspectral remote sensing data are very useful for crop feature extraction and classification. Remote sensing data is an unstructured data and Convolutional Neural Network (CNN) can work on unstructured data efficiently. This paper presents an evaluation of CNN for crop classification using the Indian Pines standard dataset obtained from the AVIRIS sensor and the study area dataset obtained from the EO-1hyperion sensor. Optimized CNN has been tuned by training the model on different parameters. It has been compared with two classification algorithms: Deep Neural Network (DNN) and Convolutional Autoencoder. According to the test results, the proposed optimized CNN model provided better results as compared to the other two methods. CNN has given 97 ± 0.58% overall accuracy for the Indian Pines standard dataset and 78 ± 2.43% for our study area dataset.

Emotion recognition from speech plays a crucial role in enhancing human–computer interaction by enabling systems to interpret and respond to users’ emotional states. This study develops and evaluates a Speech Emotion Recognition (SER) system using three machine learning techniques; Support Vector Machines (SVM), Multilayer Perceptron (MLP), and Convolutional Neural Networks (CNNs). The system is trained and tested on the RAVDESS dataset, which contains 1,440 professionally recorded audio samples representing a wide range of emotions. Our approach involves careful preprocessing of the audio signals, extraction of key acoustic features, and comparative performance evaluation of the three models using standard metrics. Results show that each model exhibits unique strengths and limitations, with CNNs achieving the most robust feature learning and generalization. The study underscores the importance of diverse feature representation for accurate emotion classification and provides insight into how different model architectures handle emotional nuances in speech. Identified challenges such as dataset diversity, feature selection, and computational complexity are discussed, along with recommendations for future research to improve SER systems’ real-world adaptability. This work contributes to ongoing efforts toward developing emotionally aware technologies that can enhance natural human–machine communication.

The report presents an algorithm for constructing a convolutional hierarchical neural network classifier, which is a modification of the algorithm for constructing hierarchical neural network classifiers suggested before. The original algorithm was designed to exploit intrinsic class hierarchy to build a class tree with a neural network in each node classifying groups of initial classes (in a non-terminal node) or a subset of original classes (in a terminal node). The convolutional modification utilizes convolutional neural networks instead of regular fully connected networks in order to apply the model to image classification tasks. Use of class hierarchy for image classification should reduce the number of adjusted neural network parameters compared to deep convolutional neural networks, and therefore it should reduce training and inference time. In this context the algorithm may be compared with some pruning techniques. The convolutional hierarchical neural network classifier inherits some hyperparameters of a conventional hierarchical neural network classifier, like the activation threshold and the threshold by the share of voting patterns. The goal of this study was to explore different strategies of choosing these hyperparameters. To test these strategies, we used the CIFAR-10 dataset. Also, for demonstration purposes we apply the convolutional hierarchical neural network classifier to the CIFAR-100 dataset.

Hearing disabilities affect people all over the world. Nearly 360 million people have a disabling hearing loss. In the recent years, there has been a rapid increase in the number of deaf and dumb victims due to birth defects, accidents and oral diseases. Since deaf and dumb people cannot communicate with a hearing person, they have to depend on some sort of visual communication. Sign Language is not understood by the majority of hearing people. The objective of our project is to form a bridge between the hearing-impaired community and the hearing community; to further commence a two-way communication. We propose a two-way sign language translator that converts Voice to sign language and vice versa in real time. We process the video frame by frame using the OpenCV Library in Python. Furthermore, we use background subtraction method called Gaussian Mixture-based Background/Foreground Segmentation Algorithm on each frame to subtract the background. Contours of this processed image are then passed into a Deep Neural Net (DNN) to classify the frame to the corresponding written language equivalent words. For conversion from sign language to voice, we use basic Natural Language Processing and gtts to accurately conserve the grammar in the sign language. The enrolment rate and literacy among the hearing-impaired children is far below the average for the population at large. This technology will then help the normal schools better integrate the hearing-impaired community thus making education more accessible and cheaper for them. Speech Emotion Recognition (SER) is a vital aspect of human-computer interaction and emotional intelligence applications. This SER that employs Mel- frequency cepstral coefficients (MFCC) as the basis for feature extraction, followed by a Convolutional Neural Network (CNN) with a customized architecture for emotion classification. The methodology 2 involves transforming audio samples into MFCC images, enabling the application of CNNs typically used in computer vision tasks. This fusion of traditional audio analysis with deep learning techniques harnesses the power of MFCC's spectral representation and CNN's spatial pattern recognition. By modifying the CNN architecture, the model can better discern emotional cues in the MFCC images, enhancing SER performance. This strategy offers SER a viable path forward, with the ability to identify complex emotional expressions. In order to increase the accuracy of emotion recognition, it tackles the requirement for feature-rich representations and sophisticated modelling techniques. KEYWORDS: OpenCV, DNN, SER, CNN, MFCC

ecently, the widespread use of social media has increased, leading to increased concerns about cyberbullying. It has become imperative to intensify efforts and methods to detect and manage cyberbullying through social media. Arabic has recently received increasing attention to improve the classification of Arabic texts. Given the multitude of Arabic dialects used on social media platforms by Arabic speakers to express their opinions and communicate with each other, applying this approach to Arabic becomes extremely challenging due to its structural and morphological complexity. Analyzing Arabic dialects using Natural Language Processing (NLP) tools can be more challenging than Standard Arabic. In this paper, the impact of using stopword removal and derivation techniques on detecting cyberbullying in the Libyan dialect was presented. The efficiency of text classification was compared when using a Libyan dialect word list alongside pre-generated Modern Standard Arabic (MSA) lists. The texts were classified using Convolutional Neural Network (CNN) classifiers, and the experiments showed that when using Libyan dialect words, the accuracy results were 92% and 83%, and when using only Standard Arabic stop words, the accuracy results were dropped to 91% and 77%. Based on these results, the higher accuracy was obtained when using the presented stop words list which it is specific to the Libyan dialect, and they had a positive impact on the results, better than Standard Arabic stop words.

Conventional geographic object-based image analysis (GEOBIA) land cover classification methods by using very high resolution images are hardly applicable due to their complex ground truth and manually selected features, while convolutional neural networks (CNNs) with many hidden layers provide the possibility of extracting deep features from very high resolution images. Compared with pixel-based CNNs, superpixel-based CNN classification, carrying on the idea of GEOBIA, is more efficient. However, superpixel-based CNNs are still problematic in terms of their processing units and accuracies. Firstly, the limitations of salt and pepper errors and low boundary adherence caused by superpixel segmentation still exist; secondly, this method uses the central point of the superpixel as the classification benchmark in identifying the category of the superpixel, which does not allow classification accuracy to be ensured. To solve such problems, this paper proposes a region-based majority voting CNN which combines the idea of GEOBIA and the deep learning technique. Firstly, training data was manually labeled and trained; secondly, images were segmented under multiresolution and the segmented regions were taken as basic processing units; then, point voters were generated within each segmented region and the perceptive fields of points voters were put into the multi-scale CNN to determine their categories. Eventually, the final category of each region was determined in the region majority voting system. The experiments and analyses indicate the following: 1. region-based majority voting CNNs can fully utilize their exclusive nature to extract abstract deep features from images; 2. compared with the pixel-based CNN and superpixel-based CNN, the region-based majority voting CNN is not only efficient but also capable of keeping better segmentation accuracy and boundary fit; 3. to a certain extent, region-based majority voting CNNs reduce the impact of the scale effect upon large objects; and 4. multi-scales containing small scales are more applicable for very high resolution image classification than the single scale.

Power data auditing is the cornerstone of a reliable and efficient modern power system. Various deep learning models have been successfully applied to fraud detection in power audit data. However, most of these methods rely on manual trial-and-error and expert knowledge to design the neural architectures and hyper-parameters. To address this limitation, this paper proposes an innovative automated deep learning approach for fraud detection model design using genetic algorithm (GA)-based neural architecture search (NAS), termed NAS-FD. In NAS-FD, convolutional neural network (CNN) is employed as the primary detection model, leveraging its strong data learning and feature extraction capabilities. First, an effective encoding scheme is developed to represent the nueral architectures and hyper-parameters of CNN, as these parameters significantly influence the detection performance. Then, considering detection performance as the objective function, well-designed GA-based evolutionary operations are implemented to optimize the neural architectures and hyper-parameters of CNN, obtaining the optimized CNN. The detection performance of the proposed NAS-FD method is validated using an electricity theft dataset from the power auditing domain. Experimental results demonstrate that NAS-FD achieves superior detection performance compared with manually designed deep learning models in terms of four performance indices including accuracy, precision, recall, and F1-score.

White blood cells (WBCs) are the most diverse types of cells observed in the healing process of injured skeletal muscles. In the recovery process, WBCs exhibit a dynamic cellular response and undergo multiple changes of the protein expression. The progress of healing can be analyzed by the number of WBCs or by the number of specific proteins observed in light microscopy images obtained at different time points after injury. We propose a deep learning quantification and analysis system called DeepQuantify to analyze WBCs in light microscopy images of uninjured and injured muscles of female mice. The DeepQuantify system features in segmentation using the localized iterative Otsu's thresholding method, masking postprocessing, and classification of WBCs with a convolutional neural network (CNN) classifier to achieve a high accuracy and a low manpower cost. The proposed two-layer CNN classifier designed based on the optimization hypothesis is evaluated and compared with other CNN classifiers. The DeepQuantify system adopting these CNN classifiers is evaluated for quantifying CD68-positive macrophages and 7/4-positive neutrophils and compared with the state-of-the-art deep learning segmentation architectures. DeepQuantify achieves an accuracy of 90.64% and 89.31% for CD68-positive macrophages and 7/4-positive neutrophils, respectively. The DeepQuantify system employing the proposed two-layer CNN architecture achieves better performance than those deep segmentation architectures. The quantitative analysis of two protein dynamics during muscle recovery is also presented.

Unsupervised feature selection and NMF de-noising for robust Speech Emotion Recognition

Emotion recognition plays a pivotal role in advancing Human-Computer Interaction (HCI) by enabling systems to understand and respond to human emotional states. Among various modalities, Speech Emotion Recognition (SER) stands out as a non-invasive, cost-effective, and temporally efficient approach for detecting emotions. This study leverages the RAVDESS dataset, which provides high-quality audio recordings for emotion classification. The proposed methodology involves preprocessing audio signals to remove noise, extracting temporal and spectral features in both time and frequency domains, and implementing machine learning models for multi-class emotion classification. The study evaluates and compares the performance of several classification models, including Random Forest (RF), Multilayer Perceptron (MLP), Support Vector Machine (SVM), Convolutional Neural Networks (CNN), and Decision Trees (DT). Experimental results demonstrate promising accuracy levels, highlighting the potential of these machine learning techniques in SER. This research contributes to the integration of SER within Brain-Computer Interfaces (BCI) and other emotion-aware applications, paving the way for enhanced interactive systems. Keywords – Speech Emotion Recognition -Machine Learning - Random Forest -Multilayer Perceptron - Support Vector Machine - Convolutional Neural Networks - Decision Tree - RAVDESS Dataset