Abstract :

The Rapid Advancement Of Artificial Intelligence (AI) And Deep Learning Technologies Has Led To The Development Of Highly Realistic Synthetic Audio, Commonly Known As Audio Deepfakes. These Manipulated Audio Recordings Can Imitate Human Voices With Remarkable Accuracy, Creating Serious Concerns Related To Misinformation, Identity Theft, Cyber Fraud, And Digital Security. Detecting Such Forged Audio Has Become An Important Challenge In The Fields Of Cybersecurity, Digital Forensics, And Media Authentication. This Research Presents An Intelligent Audio Deepfake Detection System Using Deep Learning Techniques To Classify Audio Samples As Genuine Or Fake. The Proposed System Utilizes Audio Preprocessing And Feature Extraction Methods Such As MelFrequency Cepstral Coefficients (MFCCs), Spectrogram Analysis, And Frequency-based Representations To Capture Hidden Characteristics Of Speech Signals. Deep Learning Models Including Convolutional Neural Networks (CNNs) And Long Short-Term Memory (LSTM) Networks Are Employed To Learn Discriminative Patterns Between Authentic And Synthetic Audio Samples. The Model Is Trained And Evaluated Using A Dataset Containing Both Real And AI-generated Audio Recordings. Performance Evaluation Is Carried Out Using Metrics Such As Accuracy, Precision, Recall, And F1-score. Experimental Results Demonstrate That The Proposed Deep Learning Framework Achieves High Detection Accuracy And Effectively Identifies Manipulated Audio Even Under Challenging Conditions. The Study Also Highlights The Importance Of Robust Feature Extraction And Temporal Pattern Analysis In Improving Deepfake Detection Performance. The Proposed System Can Be Applied In Real-world Domains Such As Media Verification, Voice Authentication, Digital Forensics, Banking Security, And Cybercrime Prevention. This Work Contributes Toward Building Reliable Automated Solutions For Combating The Growing Threat Of AI-generated Audio Deepfakes And Enhancing Trust In Digital Communication Systems.

Published:

15-2-2026

Issue:

Vol. 26 No. 2 (2026)

Page Nos:

278-284

Section:

Articles

License:

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

How to Cite