Lip2Speech: lightweight multi-speaker speech reconstruction with Gabor features

Computer Science

• Gabor feature 100%
• Convolutional Neural Network 28%
• Feature Extraction 28%
• Neural Network Model 14%
• Deep Learning Model 14%
• Generative Adversarial Networks 14%
• Long Short-Term Memory Network 14%
• Superior Performance 14%
• Region Feature 14%
• Processing Capability 14%

Engineering

• Supplementary Information 25%
• Speech Signal 12%
• Mouth Region 12%