ETRI-Knowledge Sharing Plaform

Artificial Intelligence System Automatic Translation Reduction Rate Neural Networks Self-Improving Automatic Speech Recognition(ASR) Multi-Domain Deep Neural Network(DNN) Acoustic Modeling Decision Tree(DT) Text Corpus Translation Model Language Barrier Mobile Devices Speech Synthesis Speech Recognition System N-Gram Deep Learning(DL) Speech-To-Speech Translation System Preliminary Study Model Parameter Error Reduction Barrier-Free Model-Based Basic Functions Cepstral Features Diagnosis System Excellent Performance Energy Operator Filtering Process Speech-To-Text(STT) Search Results Linear Transformation Online Services Parameter Reduction Performance Levels Convolution Neural Network(CNN) High-Quality User-Friendly User Satisfaction Text-To-Speech System Gradient Boosting Two-Stage Speech Recognition Technology Simple Structure Adaptive Parameter Code-Switching Construction Procedure Contextual Information Data Sets Higher-Order Statistics Lattice Network Tree-Based End-To-End Speech Recognition European Languages End Point Detection(EPD) Freely Available Training Data Time Information Structured Classification Speech Corpus Speaker Adaptation Korean Speech Matrix Decomposition Minimum Description Length (Mdl) Criterion Machine Learning Parallel Data Machine Translation(MT) Divergence Measure Arabic Speech Recognition Break Prediction Break Strength Computer Interface Classification Algorithm Hidden Markov Model Robust Speech Recognition Feature Set Ted Talks Voice Signal Markov Process Prior Knowledge Baseline System Consumer Devices Clean Environment Conventional Methods Classification Models Transformation Method Target Domain Feature Extraction Method Spontaneous Speech Source Language Spoken Language Translation Speech Database Singular Value Decomposition(SVD) Large Vocabulary Speech Recognition Large-Scale Linguistic Features New Method Prediction Model Affine Transformation Computationally Expensive Distance Measure DNN-HMM Recognition Model Topic Model Error Rate Translation Technology Target Language Domain Adaptation Zero-Crossing Statistical Information Large Vocabulary Continuous Speech Recognition Multi-Language Natural Language Processing Proposed Model Dnn Adaptation Test Set Training Set F-Measure Government Open Data Speech Information Sentence-Final Intonation Speech Act Lm Adaptation Kullback-Leibler (Kl) Divergence Multilingual Speech Medical Diagnosis Mobile System Non-Homogeneous Pathological Voice Human Computer Interaction(HCI) Correction Rate Chime Challenge Recognition Performance Total Cost Text Data Speech Signal Word Embedding Language Model Adaptation Mixture Components Open-Domain Dialog Optimization Methods Prediction Accuracy Quality Of Life