Audio-Visual Speech Enhancement Using Multimodal Deep Convolutional Neural Networks

Keyphrases

• Speech Enhancement 100%
• Deep Convolutional Neural Network (deep CNN) 100%
• Single-channel Speech Enhancement 100%
• Audio 62%
• Convolutional Neural Network 37%
• Enhancement Model 37%
• Visual Information 25%
• Output Layer 25%
• Enhanced Speech 25%
• Audio Information 25%
• Neural Network 12%
• Superior Performance 12%
• Network Model 12%
• Noise Reduction 12%
• Reconstructed Image 12%
• Audio Stream 12%
• Speech Signal 12%
• Audio Signal 12%
• Encoder-decoder 12%
• Audio Data 12%
• Backpropagation 12%
• Video Encoder 12%
• Enhancement Process 12%
• Enhancement Approaches 12%
• Joint Network 12%
• Visual Data 12%
• Unified Network 12%
• Multi-task Learning Framework 12%
• Visual Signals 12%
• Multimodal Learning 12%
• Audio-only 12%
• Visual Stream 12%

Computer Science

• Speech Enhancement 100%
• Deep Convolutional Neural Networks 100%
• Convolutional Neural Network 60%
• Visual Informations 20%
• Superior Performance 10%
• Learning Framework 10%
• Reconstructed Image 10%
• Multitask Learning 10%
• Secondary Task 10%
• Multimodal Learning 10%