Sentiment Analysis of Nepali COVID19 Tweets Using NB, SVM AND LSTM
Volume-3 | Issue-3
Deniable Authentication Encryption for Privacy Protection using Blockchain
Volume-3 | Issue-3
Blockchain-Enabled Federated Learning on Kubernetes for Air Quality Prediction Applications
Volume-3 | Issue-3
Smart Fashion: A Review of AI Applications in Virtual Try-On & Fashion Synthesis
Volume-3 | Issue-4
Hybrid Parallel Image Processing Algorithm for Binary Images with Image Thinning Technique
Volume-3 | Issue-3
Smart Medical Nursing Care Unit based on Internet of Things for Emergency Healthcare
Volume-3 | Issue-4
QoS-aware Virtual Machine (VM) for Optimal Resource Utilization and Energy Conservation
Volume-3 | Issue-3
Probabilistic Neural Network based Managing Algorithm for Building Automation System
Volume-3 | Issue-4
Fusion based Feature Extraction Analysis of ECG Signal Interpretation - A Systematic Approach
Volume-3 | Issue-1
Multi-scale CNN Approach for Accurate Detection of Underwater Static Fish Image
Volume-3 | Issue-3
Real Time Anomaly Detection Techniques Using PySpark Frame Work
Volume-2 | Issue-1
Deniable Authentication Encryption for Privacy Protection using Blockchain
Volume-3 | Issue-3
Smart Fashion: A Review of AI Applications in Virtual Try-On & Fashion Synthesis
Volume-3 | Issue-4
Sentiment Analysis of Nepali COVID19 Tweets Using NB, SVM AND LSTM
Volume-3 | Issue-3
Audio Tagging Using CNN Based Audio Neural Networks for Massive Data Processing
Volume-3 | Issue-4
Frontiers of AI beyond 2030: Novel Perspectives
Volume-4 | Issue-4
Smart Medical Nursing Care Unit based on Internet of Things for Emergency Healthcare
Volume-3 | Issue-4
Early Stage Detection of Crack in Glasses by Hybrid CNN Transformation Approach
Volume-3 | Issue-4
ARTIFICIAL INTELLIGENCE APPLICATION IN SMART WAREHOUSING ENVIRONMENT FOR AUTOMATED LOGISTICS
Volume-1 | Issue-2
Deep Convolution Neural Network Model for Credit-Card Fraud Detection and Alert
Volume-3 | Issue-2
Volume - 4 | Issue - 3 | september 2022
Published
15 September, 2022
One of the biggest threats in the speaker verification system is that of fake audio attacks. Over the years several detection approaches have been introduced that were designed to provide efficient and spoof-proof data-specific scenarios. However, the speaker verification system is still exposed to fake audio threats. Hence to address this issue, several authors have proposed methodologies to retrain and finetune the input data. The drawback with retraining and fine-tuning is that retraining requires high computation resources and time while fine-tuning results in degradation of performance. Moreover, in certain situations, the previous data becomes unavailable and cannot be accessed immediately. In this paper, we have proposed a solution that detects fake without continual-learning based methods and fake detection without forgetting in order to develop a new model which is capable of detecting spoofing attacks in an incremental fashion. In order to retain original model memory, knowledge distillation loss is introduced. In several scenarios, the distribution of genuine voice is said to be very consistent. In several scenarios, there is consistency in distribution of genuine voice hence a similarity loss is embedded additionally to perform a positive sample alignment. The output of the proposed work indicates an error rate reduction of up to 80% as observed and recorded.
KeywordsDeep learning fake without forgetting data training continuous learning fake audio detection
Full Article PDF Download Article PDF