IEEE

Channel–Spatial Aligned Global Knowledge Distillation for Underwater Acoustic Target Recognition

Abstract: Knowledge distillation (KD) is a predominant technique to streamline deep-learning-based recognition models for practical underwater deployments. However, existing KD methods for underwater ...
GitHub

AudioFuse: Unified Spectral-Temporal Learning via a Hybrid ViT-1D CNN Architecture for Biomedical Audio Classification

Official TensorFlow/Keras implementation for the paper: "AudioFuse: Unified Spectral-Temporal Learning via a Hybrid ViT-1D CNN Architecture for Phonocardiogram Classification" Summary: The automatic ...
IEEE

Mixed Intrusion Vibration Signal Recognition Based on Fiber Bragg Grating and RL-KAN-CNN

Abstract: Fiber Bragg Grating (FBG) sensing systems have demonstrated strong potential for distributed vibration monitoring, yet recognizing mixed intrusion events remains challenging due to the ...
IEEE

StealthPhase: Towards a Stealthy Backdoor Attack Against Speaker Recognition

Abstract: Speaker recognition systems (SRS) play a vital role in identity authentication. At the same time, researchers have found that these systems are highly vulnerable to backdoor attacks, where ...
IEEE

Emotion Recognition from Speech Signals by Mel-Spectrogram and a CNN-RNN

Abstract: Speech emotion recognition (SER) in health applications can offer several benefits by providing insights into the emotional well-being of individuals. In this work, we propose a method for ...
IEEE

Federated Quantum Kernel-Based Long Short-Term Memory for Human Activity Recognition

Abstract: In this work, we introduce the Federated Quantum Kernel-Based Long Short-term Memory (Fed-QK-LSTM) framework, integrating the quantum kernel methods and Long Shortterm Memory into federated ...
IEEE

Intelligent Fault Diagnosis of Motor Bearings Based on Dual-Branch Depthwise Separable CNN with Acoustic-Vibration Signal Fusion

Abstract: The intelligent diagnosis of motor bearings under complex working conditions presents significant challenges, including insufficient feature extraction, limited reliability of ...