Hybrid CNN-RNN Architecture with MFCC-LFCC Features for Audio Deepfake Detection
DOI:
https://doi.org/10.63158/journalisi.v8i4.1723Keywords:
Audio Deepfake Detection, Hybrid CNN-RNN, MFCC, LFCC, Audio Spoofing Detection, Deepfake Audio ForensicsAbstract
The proliferation of sophisticated audio deepfake technology poses a significant threat to digital voice authentication and forensic verification systems. This research addresses this challenge by developing and evaluating a lightweight hybrid Convolutional Neural Network and Recurrent Neural Network (CNN-RNN) architecture for audio deepfake detection. The proposed model integrates a CNN for spatial feature extraction with a bidirectional RNN for temporal dependency modeling, utilizing an early vertically fused feature set of Mel-Frequency Cepstral Coefficients (MFCC) and Linear Frequency Cepstral Coefficients (LFCC) stabilized via utterance-level Cepstral Mean and Variance Normalization (CMVN). We assessed the proposed framework on the official ASVspoof 2019 Logical Access (LA) evaluation benchmark dataset (71,237 trials). Comprehensive evaluation on the official evaluation set demonstrated promising performance within the evaluated benchmark, achieving a Global Equal Error Rate (EER) of 8.24% and an Area Under the Receiver Operating Characteristic (ROC-AUC) of 0.9680, while maintaining an internal validation EER of 0.33% on known attacks. While showing high sensitivity to bona fide speech and robust resilience against advanced Neural Text-to-Speech synthesis (EER < 0.25% for A07–A10), the framework exhibits notable vulnerability to phase-preserving voice conversion attacks. Consequently, without real-world forensic operational testing, the model serves as an initial diagnostic screening approach rather than a fully operational forensic solution, and still requires further cross-repository and noisy-condition validation.
Downloads
References
[1] N. Chakravarty and M. Dua, "A lightweight feature extraction technique for deepfake audio detection," Multimed. Tools Appl., vol. 83, no. 26, pp. 67443-67467, 2024. doi: 10.1007/s11042-024-18217-9.
[2] H. Radwan, A. Y. Mahmoud, M. A. El-Moneim, and H. M. El-Bakry, "Siam-CNNNet: A novel fusion of Siamese Network and Convolutional Neural Networks based on Mel-Frequency Cepstral Coefficients for audio deepfake detection," in Lecture Notes on Data Engineering and Communications Technologies, vol. 233, Springer, 2024, pp. 193-202. doi: 10.1007/978-3-031-77299-3_19.
[3] K. Duhan and A. Kajal, "Deep learning-based techniques for identification of audio deepfake with open issues: A meta-analysis," SSRG Int. J. Electr. Electron. Eng., vol. 12, no. 3, pp. 36-44, 2025. doi: 10.14445/23488379/IJEEE-V12I3P104.
[4] J. Yi, R. Fu, J. Tao, S. Nie, H. Ma, and C. Wang, "ADD 2022: The first audio deep synthesis detection challenge," in Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2022), 2022, pp. 9216-9220. doi: 10.1109/ICASSP43922.2022.9746939.
[5] P. Kawa, M. Plata, and P. Syga, “Attack agnostic dataset: towards generalization and stabilization of audio deepfake detection,” in Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH, 2022, pp. 4023–4027. doi: 10.21437/Interspeech.2022-10078.
[6] Y. Guo, H. Huang, X. Chen, H. Zhao, and Y. Wang, “Audio deepfake detection with self-supervised wavlm and multi-fusion attentive classifier,” in ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, 2024, pp. 12702–12706. doi: 10.1109/ICASSP48485.2024.10447923.
[7] J. Khochare, C. Joshi, B. Yenarkar, S. Suratkar, and F. Kazi, “A deep learning framework for audio deepfake detection,” Arab. J. Sci. Eng., vol. 47, no. 3, pp. 3447–3458, 2022, doi: 10.1007/s13369-021-06297-w.
[8] Q. Zhang, S. Wen, and T. Hu, “Audio deepfake detection with self-supervised XLS-R and SLS classifier,” in MM 2024 - Proceedings of the 32nd ACM International Conference on Multimedia, 2024, pp. 6765–6773. doi: 10.1145/3664647.3681345.
[9] P. Kawa, M. Plata, and P. Syga, “Defense against adversarial attacks on audio deepfake detection,” in Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH, 2023, pp. 5276–5280. doi: 10.21437/Interspeech.2023-409.
[10] R. Yan, C. Wen, S. Zhou, T. Guo, W. Zou, and X. Li, “Audio deepfake detection system with neural stitching for add 2022,” in ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, 2022, pp. 9226–9230. doi: 10.1109/ICASSP43922.2022.9746820.
[11] R. Liu, J. Zhang, and G. Gao, “Multi-space channel representation learning for mono-to-binaural conversion based audio deepfake detection,” Inf. Fusion, vol. 105, 2024, doi: 10.1016/j.inffus.2024.102257.
[12] R. Liu, J. Zhang, G. Gao, and H. Li, “Betray oneself: a novel audio deepfake detection model via mono-to-stereo conversion,” in Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH, 2023, pp. 3999–4003. doi: 10.21437/Interspeech.2023-2335.
[13] H.-S. Shin, J. Heo, J.-H. Kim, C.-Y. Lim, W. Kim, and H.-J. Yu, “HM-Conformer: A conformer-based audio deepfake detection system with hierarchical pooling and multi-level classification token aggregation methods,” in ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, 2024, pp. 10581–10585. doi: 10.1109/ICASSP48485.2024.10448453.
[14] V. K. Sharma, R. Garg, and Q. Caudron, “A systematic literature review on deepfake detection techniques,” Multimed. Tools Appl., 2024, doi: 10.1007/s11042-024-19906-1.
[15] I. Amerini et al., “Deepfake media forensics: status and future challenges,” J. Imaging, vol. 11, no. 3, 2025, doi: 10.3390/jimaging11030073.
[16] Z. Khanjani, G. Watson, and V. P. Janeja, “Audio deepfakes: a survey,” Front. Big Data, vol. 5, 2023, doi: 10.3389/fdata.2022.1001063.
[17] J. Li, L. Li, M. Luo, X. Wang, S. Qiao, and Y. Zhou, “Multi-grained backend fusion for manipulation region location of partially fake audio,” in CEUR Workshop Proceedings, 2023, pp. 43–48.
[18] Y. Xu, B. Li, S. Tan, and J. Huang, “Research progress on speech deepfake and its detection techniques,” J. Image Graph., vol. 29, no. 8, pp. 2236–2268, 2024, doi: 10.11834/jig.230476.
[19] Y. Xie et al., “Generalized source tracing: detecting novel audio deepfake algorithm with real emphasis and fake dispersion strategy,” pp. 4833–4837, 2024, doi: 10.21437/interspeech.2024-254.
[20] R. Anagha, A. Arya, V. H. Narayan, S. Abhishek, and T. Anjali, “Audio deepfake detection using deep learning,” in Proceedings of the 2023 12th International Conference on System Modeling and Advancement in Research Trends, SMART 2023, 2023, pp. 176–181. doi: 10.1109/SMART59791.2023.10428163.
[21] J. M. Martín-Doñas and A. Álvarez, “The vicomtech audio deepfake detection system based on wav2vec2 for the 2022 add challenge,” in ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, 2022, pp. 7937–7941. doi: 10.1109/ICASSP43922.2022.9747768.
[22] A. Khan, K. M. Malik, and S. Nawaz, “Frame-to-utterance convergence: a spectra-temporal approach for unified spoofing detection,” in ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, 2024, pp. 10761–10765. doi: 10.1109/ICASSP48485.2024.10447500.
[23] J. Frank and L. Schönherr, “WaveFake: A data set to facilitate audio deepfake detection,” Adv. Neural Inf. Process. Syst., no. NeurIPS, 2021.
[24] K. Thakral, H. Agarwal, K. Naraya, S. Mittal, M. Vatsa, and R. Singh, “DeePhyNet: towards detecting phylogeny in deepfakes,” IEEE Trans. Biometrics, Behav. Identity Sci., 2024, doi: 10.1109/TBIOM.2024.3487482.
[25] J. Yang, R. K. Das, and H. Li, “Significance of subband features for synthetic speech detection,” IEEE Trans. Inf. Forensics Secur., vol. 15, pp. 2160–2170, 2020, doi: 10.1109/TIFS.2019.2956589.
[26] Y. Zhang, J. Lu, Z. Li, Z. Shang, W. Wang, and P. Zhang, “Improving the robustness of deepfake audio detection through confidence calibration,” in CEUR Workshop Proceedings, 2023, pp. 70–75.
[27] S. A. Momu, R. R. Siddiqui, S. S. Shanto, and Z. Ahmed, “A comprehensive approach to deepfake audio detection: using feature fusion and deep learning,” in 2024 27th International Conference on Computer and Information Technology, ICCIT 2024 - Proceedings, 2024, pp. 351–356. doi: 10.1109/ICCIT64611.2024.11022599.
[28] K. Schäfer and M. Steinebach, “MFCC vs. LFCC for audio deepfake detection: the role of delta features and input length,” in European Signal Processing Conference, 2025, pp. 576–580. doi: 10.23919/EUSIPCO63237.2025.11226775.
[29] X. Wang et al., “ASVspoof 2019: A large-scale public database of synthesized, converted and replayed speech,” 2020, Accessed: Jun. 25, 2026. [Online]. Available: https://www.asvspoof.org/index2019.html
[30] Z. K. Abdul and A. K. Al-Talabani, “Mel frequency cepstral coefficient and its applications: a review,” IEEE Access, vol. 10, pp. 122136–122158, 2022, doi: 10.1109/ACCESS.2022.3223444.
[31] H. Tak, J. Patino, M. Todisco, A. Nautsch, N. Evans, and A. Larcher, “End-to-end anti-spoofing with Rawnet2,” in ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, 2021, pp. 6369–6373. doi: 10.1109/ICASSP39728.2021.9414234.
[32] K. Z. Mon, K. Galajit, C. O. Mawalim, J. Karnjana, T. Isshiki, and P. Aimmanee, “Spoof detection using voice contribution on LFCC features and ResNet-34,” in 18th International Conference on Artificial Intelligence and Natural Language Processing and International Conference on Artificial Intelligence and Internet of Things, iSAI-NLP 2023, 2023. doi: 10.1109/iSAI-NLP60301.2023.10354625.
[33] K. S. Krishnan and K. S. Krishnan, “MFAAN: Unveiling audio deepfakes with a multi-feature authenticity network,” in 2023 9th International Conference on Signal Processing and Communication, ICSC 2023, 2023, pp. 150–155. doi: 10.1109/ICSC60394.2023.10441405.
[34] R. Mahyavanshi, C. V Mahesh Reddy, A. J. Shah, and H. A. Patil, “Teager energy cepstral coefficients for audio deepfake detection,” in APSIPA ASC 2024 - Asia Pacific Signal and Information Processing Association Annual Summit and Conference 2024, 2024. doi: 10.1109/APSIPAASC63619.2025.10848893.
[35] J. Zhou and L. Yang, “Research on audio scene classification method based on deep learning technology in sound processing,” in ACM International Conference Proceeding Series, 2024, pp. 664–668. doi: 10.1145/3675417.3675527.
[36] L. Z. Yong and H. Nugroho, “Acoustic anomaly detection of mechanical failure: time-distributed CNN-RNN deep learning models,” in Lecture Notes in Electrical Engineering, 2022, pp. 662–672. doi: 10.1007/978-981-19-3923-5_57.
[37] L. Pham, P. Lam, T. Nguyen, H. Nguyen, and A. Schindler, “Deepfake audio detection using spectrogram-based feature and ensemble of deep learning models,” in IEEE 5th International Symposium on the Internet of Sounds, IS2 2024, 2024. doi: 10.1109/IS262782.2024.10704095.
[38] K. Li, X.-M. Zeng, J.-T. Zhang, and Y. Song, “Convolutional recurrent neural network and multitask learning for manipulation region location,” in CEUR Workshop Proceedings, 2023, pp. 18–22.
[39] H. Tak, J. Patino, A. Nautsch, N. Evans, and M. Todisco, “Spoofing attack detection using the non-linear fusion of sub-band classifiers,” in Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH, 2020, pp. 1106–1110. doi: 10.21437/Interspeech.2020-1844.
[40] R. Lomte, P. Singhal, S. Patil, Z. Shaikh, and P. Agarwal, “A multi-scale residual network with hierarchical feature fusion for robust audio deepfake detection,” in Lecture Notes in Networks and Systems, 2026, pp. 449–466. doi: 10.1007/978-3-032-13803-3_36.
[41] K. Galajit et al., “ThaiSpoof: A database for spoof detection in Thai language,” in 18th International Conference on Artificial Intelligence and Natural Language Processing and International Conference on Artificial Intelligence and Internet of Things, iSAI-NLP 2023, 2023. doi: 10.1109/iSAI-NLP60301.2023.10354956.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Journal of Information Systems and Informatics
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors Declaration
- The Authors certify that they have read, understood, and agreed to the Journal of Information Systems and Informatics (JournalISI) submission guidelines, policies, and submission declaration. The submission has been prepared using the provided template.
- The Authors certify that all authors have approved the publication of this manuscript and that there is no conflict of interest.
- The Authors confirm that the manuscript is their original work, has not received prior publication, is not under consideration for publication elsewhere, and has not been previously published.
- The Authors confirm that all authors listed on the title page have contributed significantly to the work, have read the manuscript, attest to the validity and legitimacy of the data and its interpretation, and agree to its submission.
- The Authors confirm that the manuscript is not copied from or plagiarized from any other published work.
- The Authors declare that the manuscript will not be submitted for publication in any other journal or magazine until a decision is made by the journal editors.
- If the manuscript is finally accepted for publication, the Authors confirm that they will either proceed with publication immediately or withdraw the manuscript in accordance with the journal’s withdrawal policies.
- The Authors agree that, upon publication of the manuscript in this journal, they transfer copyright or assign exclusive rights to the publisher, including commercial rights
