1 August 2025 in Indonesian Journal of Electrical Engineering and Computer Science
Tajweed is the science of reciting the Holy Quran, focusing on the clarity and correctness of recitation. This paper aims to accurately detect the spoken Tajweed rules applied during Quranic recitation, providing a well-structured Tajweed rules database for further analysis, Tajweed learning, and the training of advanced classification models. The main contribution of this work is to identify a high-accuracy approach for Tajweed rules detection and analysis. An improved template matching approach is introduced to enhance detection accuracy by matching the Quranic verse audio file with multiple speech patterns of a specific rule and selecting the best match. The Quranic audio file is segmented into smaller patterns by finding the correlation between the adjacent audio frames. Then, the template matching is applied to these segmented patterns to identify the best-matching ones. The template matching technique relies on a Tajweed database of 487 patterns of the Madd, Noon Sakinah, Tanween, and Meem Sakinah rules. An overall detection accuracy of 97.1% is achieved, and the Tajweed-pattern database is expanded to include the newly detected rules, increasing their total count to 2,583. Furthermore, an application based on the detected rules in this study was developed to enhance the performance of new Tajweed learners.
1 September 2024 in International Journal of Advances in Applied Sciences
The most frequent type of cancer among women is breast cancer. Artificial intelligence (AI) researchers are developing automated systems to assist in the detection and classification of breast cancer. This study explores machine learning (ML) and deep learning (DL) as two AI methods for identifying benign and malignant breast tumors in ultrasound images. The investigation assesses the performance of various computer-aided detection and diagnosis (CAD) systems, which utilize either handcrafted features or deep features extracted from DL models. Furthermore, three models for CAD deep learning-based systems were implemented using convolutional neural networks (CNN), convolutional autoencoders (CAE), and deep features with CNN models, and compared with three traditional ML models based on handcrafted (texture) features. The results indicate that the deep features of the CNN model are promising, achieving a mean accuracy of 95% with a standard deviation of 1.1%.
21 July 2025 in Menoufia Journal of Electronic Engineering Research
This paper presents the development and implementation of an Automatic Speech Recognition (ASR)-based Audio Dictionary Database for Quranic words, integrating modern speech processing and recognition techniques. The system utilizes speech feature extraction, pattern recognition, and audio matching to detect and log Quranic words from recitations. A computational approach extracts all 19,058 unique dictionary words from the Quranic text, and a dedicated speech processing application has been developed to detect these words in Quranic recitations. The application employs enhanced speech pattern-matching techniques, successfully identifying all dictionary words in Chapter 30 (1,272 words) and spotting 22 Quranic words across the entire Quran in 373 occurrences, achieving 95% accuracy. The Audio Dictionary of Quranic Words is a novel contribution that addresses the shortage of publicly available datasets for researchers. This dataset can be used to support Tajweed instruction, advance Arabic speech technology, and serve as a valuable resource for linguistic research, ASR development, and speech-based AI applications.