ETRI-Knowledge Sharing Plaform

Acoustic Model Self-Improving Compensation Method Artificial Intelligence System Support Vectormachine(SVM) Log Likelihood Ratio(LLR) Confidence Measure Word-Level Isolated Word Recognition Embedded Devices Low-Resource Truncated Singular Value Decomposition Barrier-Free Frame-Level Low Correlation Nonlinear Manifold Objective Function Parameter Reduction Universal Background Model Within-Class Covariance Normalization Automatic Translation Position Information Distance Method Model Parameter Non-Native Speech Similar Words Reduction Technique Training Data Distance Algorithm Equal Error Rate Vector Based Correct Position System Performance Channel Effect Computing Technology Factorization Approach Robust Speech Recognition Speech Recognition System Deep Neural Network(DNN) I-Vector Matrix Factorization Phone Recognition Cosine Distance Channel Compensation Linear Discriminant Analysis(LDA) Manifold Structure Total Variability Best Performance Distance Measure Word Similarity Frequency Ratio Low-Rank Matrix Non-Similar Acoustic Features Euclidean Norm Knowledge-Based Target Matrix Computational Complexity Cognitive Computing Language Barrier Manifold Regularized Relative Reduction