Spatial reconstructed local attention Res2Net with F0 subband for fake speech detection

C Fan, J Xue, J Tao, J Yi, C Wang, C Zheng, Z Lv - Neural Networks, 2024 - Elsevier
The rhythm of bonafide speech is often difficult to replicate, which causes that the
fundamental frequency (F0) of synthetic speech is significantly different from that of real …

An improved feature extraction for Hindi language audio impersonation attack detection

N Chakravarty, M Dua - Multimedia Tools and Applications, 2024 - Springer
Audio impersonation attacks offer a substantial risk to voice-based authentication systems
and various speech recognition applications. Hence, there is a requirement for robust …

The Three-Stage Hierarchical Logistic Model Controlling Personalised Playback of Audio Information for Intelligent Tutoring Systems

AN Varnavsky - IEEE Transactions on Learning Technologies, 2024 - ieeexplore.ieee.org
The most critical parameter of audio and video information output is the playback speed,
which affects many viewing or listening metrics, including when learning using tutoring …

An explainable deepfake of speech detection method with spectrograms and waveforms

N Yu, L Chen, T Leng, Z Chen, X Yi - Journal of Information Security and …, 2024 - Elsevier
Research on deepfake techniques for speech is crucial for combatting the spread of fake
information, safeguarding public privacy, and advancing forensic techniques. However, the …

Artificial Intelligence to Combat Audio Fraud: A Flask-Deployed Hybrid Deep Learning System

S Solanki, RN Ravikumar, T Devataval… - 2024 IEEE 12th …, 2024 - ieeexplore.ieee.org
As the use of synthetic communication is growing widely spread, the ability to accurately
identify counterfeit spoken words is now more important than ever. In this study, we have …

Voice Analysis for Detecting Artificial Speech and Predicting Gender and Age: A Literature Survey

R Bharadwaj, R Dugad, A Gile, G Patil… - 2023 7th International …, 2023 - ieeexplore.ieee.org
This article surveys the literature on voice analysis. The present invention relates to a
method and system for analyzing voice data to detect artificial speech then demodulate it to …

Model Ekstraksi Fitur Berbasis Tekstur untuk Identifikasi Keaslian Objek

FT Kurniati - Jurnal Sistem dan Informatika (JSI), 2024 - jsi.stikom-bali.ac.id
Identifikasi keaslian objek merupakan aspek penting dalam berbagai sektor, termasuk
keamanan dan perdagangan, guna mencegah kerugian finansial dan reputasi akibat …

Voice Analysis for Detecting Artificial Speech and Predicting Gender

R Bharadwaj, R Dugad, A Gile, G Patil… - … (OTCON) on Smart …, 2024 - ieeexplore.ieee.org
This research paper presents a comprehensive voice analysis system capable of detecting
artificial speech and accurately predicting the gender and age of the speaker. Leveraging …

[PDF][PDF] Unimodal Methods for DeepFake Audio and Video Classification with Spectral Features

D Liu, J Smith, K So, A Xu, J Zhang - andorexad.github.io
The work of Durall et al [1] showed that spectral features make deepfake images highly
separable from real images. Simple classification methods thus achieve near perfect …