Discriminative subspace modeling of SNR and duration variabilities for robust speaker verification

Keyphrases

• Signal-to-noise Ratio 100%
• Robust Speaker Verification 100%
• Duration Variability 100%
• Discriminative Subspace 100%
• Subspace Model 100%
• I-vector 85%
• Probabilistic LDA 71%
• Utterance 28%
• Noise Level 28%
• Specific Information 28%
• Utterance Length 28%
• Uncertainty Propagation 14%
• Speaker Verification 14%
• Linear Combination 14%
• Vector Space 14%
• Additive Noise 14%
• Likelihood Ratio 14%
• Vector-based 14%
• Background Noise Level 14%
• Ratio Factor 14%
• Training Procedure 14%
• Speaker Recognition System 14%
• Undesirable Effects 14%
• NIST 2012 SRE 14%
• Test Utterance 14%
• Latent Variable Models 14%
• Similar Signals 14%
• Noise Channel 14%
• Variational Bayes 14%
• Bayes Algorithm 14%
• LDA Model 14%
• Subspace Training 14%

Computer Science

• Speaker Verification 100%
• Signal-to-Noise Ratio 100%
• Linear Discriminant Analysis 71%
• Verification System 14%
• Uncertainty Propagation 14%
• Undesirable Effect 14%
• Linear Combination 14%
• Background Noise 14%
• Likelihood Ratio 14%
• Latent Variable Model 14%

Engineering

• Signal-to-Noise Ratio 100%
• Noise Level 42%
• Space Vector 14%
• Additive Noise 14%
• Latent Variable Model 14%
• Linear Combination 14%
• Likelihood Ratio 14%