ETRI-Knowledge Sharing Plaform

Self-Improving Artificial Intelligence System Automatic Speech Recognition(ASR) Reduction Rate Adaptive Parameter Code-Switching Deep Learning(DL) Machine Translation(MT) Matrix Decomposition Speaker Adaptation Speech Corpus Model-Based Open Data Transformation Method Target Domain Conventional Methods Clean Environment Deep Neural Network(DNN) New Method Large-Scale Singular Value Decomposition(SVD) Spontaneous Speech Affine Transformation Computationally Expensive Large Vocabulary Continuous Speech Recognition Recognition Model Lm Adaptation Correction Rate Language Model Adaptation Open-Domain Dialog Training Data End-To-End Speech Recognition Error Reduction Performance Levels Recognition Performance Linear Transformation Online Services