Keyphrases
Joint Training
100%
End-to-end Speech Recognition
100%
Training Framework
100%
Automatic Speech Recognition
80%
Speech Enhancement
60%
Loss Function
40%
Feature Extraction
20%
Human-machine Interaction
20%
Attention-based
20%
Challenging Environments
20%
Noise Robustness
20%
Self-attention
20%
Generative Adversarial Networks
20%
Common Solution
20%
Sub-optimum
20%
Convolutional Layer
20%
Self-attention Mechanism
20%
Local Guides
20%
Overtaking
20%
Speech Recognition Performance
20%
Speech Recognition System
20%
Acoustic Conditions
20%
Sinc Convolution
20%
Novel Joint
20%
Popular
20%
Unseen
20%
Generative Adversarial Architecture
20%
Engineering
Joints (Structural Components)
100%
Front End
100%
Automatic Speech Recognition
100%
Robust Speech Recognition
100%
Speech Enhancement
60%
Loss Function
40%
Limitations
20%
Suboptimum
20%
Feature Extraction
20%
Human-Machine Interaction
20%
Computer Science
Speech Recognition
100%
Robust Speech Recognition
100%
Speech Enhancement
75%
Enhancement Module
50%
Human-Computer Interaction
25%
Generative Adversarial Networks
25%
Convolution Layer
25%
Self-Attention Mechanism
25%
Recognition Performance
25%
Back End
25%
Speech Recognition System
25%
Feature Extraction
25%
Earth and Planetary Sciences
Speech Recognition
100%
Self-Attention
33%
Pattern Recognition
16%