ETRI-Knowledge Sharing Plaform

Automatic Translation Language Barrier Self-Improving Artificial Intelligence System Barrier-Free Translation Model Deep Neural Network(DNN) Performance Levels Acoustic Model Speech Recognition System Reduction Rate Korean Speech Preliminary Study Contextual Information Error Reduction Mobile Devices Decision Tree(DT) Speech Synthesis Speech-To-Speech Translation System Text Corpus Handheld Devices N-Gram Recognition Model Training Data Automatic Speech Recognition(ASR) Model Parameter Model-Based Multi-Domain Classification Algorithm Intelligent Vehicle Peer Review Process Prediction Model Two-Stage Target Language Linear Transformation Lack Of Data Memory Consumption Noise Robust Speech Recognition System Development Spatial Cue Power Ratio Parallel Data Training Set Automatic Conversion Affine Transformation Correction Rate Distance Measure Singular Value Decomposition(SVD) Topic Model Test Cases Recognition Performance DNN-HMM Markov Process Neural Machine Translation Language Model Adaptation Break Prediction Deep Learning(DL) F-Measure Process Performance Chime Challenge Mobile System Gradient Boosting Quality Management Model Computationally Expensive Clean Environment Code-Switching Divergence Measure Open Source Structured Classification Code Validation Noisy Training Speech Database Speech Corpus Over Time Test Set Speech Recognition Technology Simple Structure Spontaneous Speech Translation Module Source Language Construction Procedure Higher-Order Statistics Hidden Markov Model(HMM) High-Quality Success Rate Korean Language Break Strength Diagnosis System Large Vocabulary Continuous Speech Recognition(LVCSR) Lattice Network Linguistic Features Open Technology Parallel Corpus Excellent Performance Cepstral Features Minimum Description Length (Mdl) Criterion Speech Information Adaptive Parameter Prior Knowledge Proper Names Infrastructure Support Government Open Data Translation Technology Feature Set Speaker Adaptation Classification Models Error Rate Prediction Accuracy Matrix Decomposition Need For Text Data Endpoint Detection Online Services Open-Domain Dialog Target Domain Convolution Neural Network(CNN) Multiple Sound Sources Large-Scale Parallel Speech Act English Language Search Results Time Information Conventional Methods Development Technologies Large Vocabulary Speech Recognition Binaural Model Zero-Crossing End-To-End Speech Recognition European Languages User Satisfaction Arabic Speech Recognition Filtering Process Gammatone Filterbank Matching Pattern Transformation Method Testing Process Word Embedding Feature Extraction Method Technology Infrastructure Tree-Based Multilingual Speech Voice Signal Design And Implementation Total Cost Text-To-Speech(TTS) Optimization Methods Cepstral Coefficients And Phrases Checking Points Multi-Language Non-Homogeneous Requirements Definition Speech Signals Lm Adaptation Pathological Voice Ted Talks Statistical Information Basic Functions Conversational Data Finite State Transducer Knowledge-Based Medical Diagnosis User-Friendly Bleu Score Data Sets Energy Operator Machine Learning Sentence-Final Intonation Consumer Devices Model Performance Kullback-Leibler (Kl) Divergence Proposed Model Project Management Viterbi Decoding Mixture Components New Method Project Implementation Project-Based Static Analysis Freely Available Baseline System