Audio Fingerprinting to Achieve Greater Accuracy and Maximum Speed with Multi-Model CNN-RNN-LSTM in Speaker Identification

International Journal of Computational and Experimental Science and Engineering, Год журнала: 2025, Номер 11(1)

Опубликована: Фев. 20, 2025

The process of matching speech data with database records is known as speaker identification. major objective this paper to find the accuracy and speed in comparison training set from RAVDESS test signal using neural network methods Convolutional Neural Network (CNN), Recurrent (RNN) along Long Short-Term Memory (LSTM) combination audio fingerprinting technique. Speech most fundamental form human communication language primary means exchange among humans. An essential component social interaction pitch tone changes are grouped together while accounting for a wide range issues. fingerprint voice was produced after background noise eliminated. Dataset multilayer perception, Audio CNN, RNN LSTM contrast results measures. machine will ultimately display gender determination relation words per second terms no epochs has been observed .and show that every classifier dataset performs faster higher accuracy.

Язык: Английский

---

Enhancing Ophthalmological Diagnoses: An Adaptive Ensemble Learning Approach Using Fundus and OCT Imaging

International Journal of Computational and Experimental Science and Engineering, Год журнала: 2024, Номер 10(4)

Опубликована: Дек. 21, 2024

This study presents an innovative Ensemble Disease Learning Algorithm (EDL) for the detection and classification of retinal diseases using fundus images. We enhance our method by incorporating deep learning techniques multi-modal imaging data, including optical coherence tomography (OCT) images alongside photographs, to provide a more comprehensive understanding pathology. The advanced EDL integrates Convolutional Neural Networks (CNNs) attention mechanisms with Capsule (CapsNet) Support Vector Machine (SVM) classifiers nuanced feature extraction classification. introduce novel ensemble adaptive weighting approach that dynamically adjusts classifier weights based on performance across disease types severity levels, significantly improving algorithm's handling complex rare cases. To model interpretability, we implement explainable AI component provides visual heatmaps most significant regions each diagnosis clinicians. evaluate enhanced large, diverse dataset encompassing multiple diseases, diabetic retinopathy, age-related macular degeneration, glaucoma, various ethnicities age groups. Our results demonstrate superior accuracy, sensitivity, specificity compared previous other state-of-the-art approaches. A prospective clinical validation assesses real-world performance. research advances automated making it robust, accurate, clinically relevant, potentially patient outcomes global eye care through early treatment planning.

Язык: Английский

---

ResDenseNet:Hybrid Convolutional Neural Network Model for Advanced Classification of Diabetic Retinopathy(DR) in Retinal Image Analysis

International Journal of Computational and Experimental Science and Engineering, Год журнала: 2024, Номер 10(4)

Опубликована: Дек. 22, 2024

Preventing vision loss in diabetic retinopathy (DR) requires early and precise detection. Although strong feature extraction is required there class imbalance the current methods, deep learning (DL) techniques have showed promise DR classification. With components from both ResNeXt DenseNet designs, a unique DL architecture for classification proposed this work. A that integrates work.To address issues classification, method channel-wise masking with an attention mechanism. The network able to learn less frequent stages because reduces influence of majority concentrates on important features. To improve interpretability confidence model's predictions, incorporation Explainable AI (XAI) approaches also covered.Our findings show suggested approach outperforms architectures, achieving better sensitivity differentiating phases at 0.82 accuracy 0.87. This shows new has improving categorization, which could result earlier diagnoses patient outcomes.

Язык: Английский

---

Subjective Clustering Approach by Edge detection for construction remodelling with dented construction materials

International Journal of Computational and Experimental Science and Engineering, Год журнала: 2024, Номер 10(4)

Опубликована: Дек. 24, 2024

An approach for Construction remodelling with subjective clustering edge detection is at hand in this evaluation. The available processes a verdict weight on comparison of trait vector c dataset by existing intellectual thinking to the crisis. proposed identifies clusters dented materials detecting edges high velocity, and area. consistent factor material choose added form load construction proper enlarge edification statistics method materials. direction value material. This leads formation convolution creation. orderly correlating civilized technique big order. However, problem information experiential be limited increase training attribute knowledge data. To conquer matter clustering, w-means expand issue intended. improves cluster data using double feature observing constraint. obtainable exemplify an upgrading removal presentation conditions correctness, compassion suggest more velocity

Язык: Английский

---