A Survey on Deep Fake Audio Detection Using Deep Learning

Sai Retish Reddy, Prakash Kumar Sarangi, Shravan Nikhil, M. A. Jabbar, Sukanya Mekala

2025

Abstract

The rise of synthetic voice generation techniques such as deepfakes and voice cloning, has raised concerns about potential misuse in areas like fraud, misinformation, and identity theft. Deep Learning has emerged as an effective solution for detecting deep fake audio. Features from audio files were extracted and analyzed using Mel-Frequency Cepstral Coefficients (MFCC) to illustrate the differences between real and fake audio. By examining patterns in the audio data, models can learn to distinguish subtle discrepancies. Results indicated that a Custom Architecture outperformed, with the VGG-16 architecture yielding the best results for MFCC image features. Using Deep Learning approaches, this paper presents an overview of deep fake audio detection methods, highlighting the importance of this study and the potential uses of Deep Learning Models in the fight against deep fake audio.

Paper Citation

in Harvard Style

Reddy S., Sarangi P., Nikhil S., Jabbar M. and Mekala S. (2025). A Survey on Deep Fake Audio Detection Using Deep Learning. In Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25; ISBN 978-989-758-777-1, SciTePress, pages 283-289. DOI: 10.5220/0013911900004919

in Bibtex Style

@conference{icrdicct`2525,
author={Sai Reddy and Prakash Sarangi and Shravan Nikhil and M. Jabbar and Sukanya Mekala},
title={A Survey on Deep Fake Audio Detection Using Deep Learning},
booktitle={Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25},
year={2025},
pages={283-289},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013911900004919},
isbn={978-989-758-777-1},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25
TI - A Survey on Deep Fake Audio Detection Using Deep Learning
SN - 978-989-758-777-1
AU - Reddy S.
AU - Sarangi P.
AU - Nikhil S.
AU - Jabbar M.
AU - Mekala S.
PY - 2025
SP - 283
EP - 289
DO - 10.5220/0013911900004919
PB - SciTePress