This paper presents an Al Quran Recitation Verification System using deep learning to assist in accurate and reliable Quranic recitation verification. The system leverages Convolutional Neural Networks (CNNs) and Long Short-Term Memory (LSTM) networks to process and recognize unique patterns in recitations, accommodating variations in accent and pronunciation. The methodology includes collecting high-quality audio recordings, preprocessing with noise reduction and normalization, and feature extraction using Mel-frequency Cepstral Coefficients (MFCCs). The CNN model, trained on this preprocessed data, accurately identifies Quranic letters and recitation patterns. The system provides automated feedback by comparing recitations against standard references. Initial results showed a test accuracy of 56.25% and a test loss of 0.97, which decreased to 33.33% accuracy and a test loss of 1.43 on a new dataset, indicating overfitting. Confusion matrices revealed a bias towards the “MIL” class, suggesting the need for improved class discrimination.