ETRI-Knowledge Sharing Plaform

Artificial Intelligence System Self-Improving Reduction Rate Automatic Speech Recognition(ASR) End-To-End Speech Recognition Error Reduction Training Data Recognition Model Speech Corpus Over Time Spontaneous Speech Korean Language Large Vocabulary Continuous Speech Recognition(LVCSR) Performance Levels Adaptive Parameter Deep Neural Network(DNN) Translation Model Speaker Adaptation Matrix Decomposition Limited Data Online Services Open-Domain Dialog Target Domain Text Normalization Data Scarcity Open Data Conventional Methods Large-Scale Performance Degradation Matching Pattern Transformation Method Machine Translation(MT) Korean Speech And Phrases Lm Adaptation Model Performance Model-Based New Method Post-Processing Clean Environment Computationally Expensive Code-Switching Translation Errors Deep Learning(DL) Language Model Adaptation Recognition Performance Singular Value Decomposition(SVD) Automatic Conversion Correction Rate Affine Transformation Linear Transformation Automatic Translation Need For