A deep joint sparse non-negative matrix factorization framework for identifying the common and subject-specific functional units of tongue motion during speech

Keyphrases

• Subject-specific 100%
• Tongue Motion 100%
• Sparse Non-negative Matrix Factorization 100%
• Joint Sparse 100%
• Tagged Magnetic Resonance Imaging 20%
• Tongue 20%
• Simulated Dataset 20%
• Comparison Method 20%
• Non-negative Matrix Factorization 20%
• Coordination Pattern 20%
• Sparse Regularization 20%
• Size Variation 20%
• Motion Data 20%
• Local muscles 20%
• Spectral Clustering 20%
• Graph Regularized 20%
• Modular Architecture 20%
• Deep Neural Network 20%
• Shrinkage Algorithm 20%
• Muscle Coordination 20%
• Intelligible Speech 20%
• Deep Learning Framework 20%
• Graph Regularization 20%
• Major Challenges 20%
• Clustering Performance 20%

Computer Science

• nonnegative matrix factorization 100%
• Functional Unit 100%
• Matrix Factorization 16%
• Regularization 16%
• Building-Blocks 16%
• Learning Framework 16%
• Spectral Clustering 16%
• Modular Architecture 16%
• Deep Neural Network 16%
• Interpretability 16%
• Deep Learning Method 16%