Adversarial Learning and Augmentation for Speaker Recognition

Keyphrases

• I-vector 100%
• Adversarial Learning 100%
• Speaker Recognition 100%
• Adversarial Data Augmentation 100%
• Probabilistic Linear Discriminant Analysis 40%
• Class Label 40%
• Generative Adversarial Networks 40%
• Learning Objectives 20%
• Low Quality Data 20%
• System Robustness 20%
• Posterior Probability 20%
• Loss Function 20%
• Utterance 20%
• Min-max Optimization 20%
• Imbalanced Data 20%
• Model Regularization 20%
• Recognition-based 20%
• Adversarial Loss 20%
• Discriminant Analysis 20%
• Cosine Similarity 20%
• Data Reconstruction 20%
• Deep Model 20%
• Data Augmentation 20%
• Multi-objective Learning 20%
• Multiple Learning 20%
• Minimax Optimization 20%
• Reconstruction Regularization 20%
• Gaussian Regularization 20%

Computer Science

• Speaker Recognition 100%
• Linear Discriminant Analysis 50%
• Generative Adversarial Networks 50%
• Discriminator 50%
• Optimization Problem 25%
• Posterior Probability 25%
• Multiobjective 25%
• Analysis Model 25%
• Regularization 25%
• Cosine Similarity 25%
• Data Augmentation 25%
• Gaussian Regularization 25%