Nenny Anggraini's Doctoral Promotion Exam: CRNN Modeling to Detect Short Length of Harakat in the Reading of Surah al-Fātiḥah
Auditorium of Prof. Dr. Suwito, MA SPs UIN Jakarta, SPs NEWS - The Graduate School (SPs) of UIN Syarif Hidayatullah Jakarta held the 1646th Doctoral Promotion Exam in the Auditorium Room of Prof. Dr. Suwito, MA on Monday, December 1, 2025 with the promovenda of Nenny Anggraini.
Nenny is a student of the Doctoral program in Islamic Studies with a concentration in Religion and Science. His dissertation, titled "Modeling of Hybrid Convolutional Recurrent Neural Network (CRNN) to Detect Short Lengths of Harakat in the Recitation of Surah al-Fātiḥah Using Speech Recognition Technology on Android Apps," offers a revolutionary solution to challenges in the teaching and learning of the Qur'ān.
The main problems raised by Nenny focus on three crucial aspects: subjectivity in determining the duration of the reading of the Qur'ān, the limitation of the annotated surah al-Fātiḥah dataset , and the lack of integration of the CRNN hybrid model with Android applications to detect the law of madd (short length). This research aims to build a model that is able to overcome this subjectivity through objective and measurable software.
This research uses the Research and Development (R&D) method with comprehensive stages ranging from preliminary studies, model development, to testing. Meanwhile, the implementation of the model into Android applications is developed using the Waterfall method, which includes needs analysis, design, implementation, and testing of the application in stages.
The novelty that is the main highlight of this dissertation is the formulation of the CRNN hybrid architecture. Nenny introduced the two-staged hybrid CRNN approach which is the first in the context of the detection of short lengths harakat al-Fātiḥah. This model is strengthened by the use of eight acoustic features at once, including MFCC, Formants, Pitch, Spectral Centroid, Spectral Bandwidth, Spectral Flatness, Onset Energy, and LPC.
The dataset used is a primary dataset developed by ourselves, ensuring accuracy and relevance to the research objectives. The highlight of this novelty is the integration of the CRNN model into an Android app, making it a practical tool for users.
The test results of the model show very impressive performance. The evaluation of the model training succeeded in recording a classification accuracy of 98%. Furthermore, model testing using 30 audio real data outside of the main dataset showed consistent accuracy, reaching 94%. These results significantly surpass previous studies.
Compared to the study of Zulkifli et al. which used the LSTM algorithm with 90% accuracy without real data testing, or the study of Shafie et al. with the Dynamic Time Warping (DTW) algorithm which produced an accuracy of 80.47% on real data, the CRNN Nenny hybrid model showed much better performance and generalization capabilities.
The ablation test revealed that MFCC (Mel-Frequency Cepstral Coefficients) was the main acoustic feature with the greatest contribution. Nonetheless, all eight acoustic feature extractions are retained to achieve maximum model performance.
The developed Android app is also extensively tested. Blackbox testing by researchers and 67 respondents ensured that all application functions were running according to specifications. The User Acceptance Test (UAT) score reached an outstanding score of 92%, and the results of the System Usability Scale (SUS) test placed the application in the "Excellent" category with an average score of 83.5.
The modeling and applications produced by Nenny are not only academic achievements, but also offer a new approach in Qur'ān education. By integrating religious science and speech recognition technology, this study provides an objective and reliable tool for detecting the truth of the long short harakat (law of madd) in the recitation of Surah al-Fātiḥah. This marks a significant step forward in the use of Artificial Intelligence (AI) to deepen religious understanding and practice.
Nenny Anggraini successfully defended her dissertation under the guidance of Prof. Dr. Yusuf Rahman, MA; Prof. Dr. Achmad Nizar Hidayanto, S.Kom, M.Kom; Prof. Husni Teja Sukmana, M.Sc, Ph.D, and tested in front of a board of examiners consisting of Prof. Dr. Zulkifli, MA; Prof. Dr. Yusuf Rahman, MA; Prof. Dr. Achmad Nizar Hidayanto, S.Kom, M.Kom; Prof. Husni Teja Sukmana, M.Sc, Ph.D; Prof. Dr. Syopiansyah Jaya Putra, M.Sis; Prof. Kusmana, MA, Ph.D; Ir. Nashrul Hakiem, S.Si, MT, Ph.D.
After paying attention to the writing of the dissertation, the comments of the examiner team and the promovenda's answers, the examiner team determined that Nenny Anggraini graduated with the predicate of Cum Laude. Nenny Anggraini is the 1646th Doctor in the field of Islamic Studies, in the doctoral program of the Graduate School of UIN Syarif Hidayatullah Jakarta. (JA)
