ETRI-Knowledge Sharing Plaform

Deep Learning(DL) Transfer Learning Automatic Speech Recognition(ASR) Deep Neural Network(DNN) Self-Training Student Learning Multi-Stage Data Augmentation Point Of Interest Learning Methods State-Of-The-Art Based System Error Reduction Latent Dirichlet Allocation (Lda) Artificial Intelligence System Latent Semantic Analysis(LSI) Learning Algorithms Signal Noise Ratio(SNR) Target Domain Tutoring System Spontaneous Speech Speech-To-Text(STT) Car Navigation System Intermediate Input Latent Semantics Computational Complexity Lda Model Assessment Method Self-Improving Reduction Rate Training Process Semi-Supervised Training Pre-Training Training Data Pre-Trained Model Speech Corpus Barrier-Free Audio Data Augmentation Techniques Behavior Strategy Conversational Speech Recognition Domain Transfer Data-Driven Approach GMM-HMM Latent Variables Layer-By-Layer(LbL) Native Language Off-The-Shelf Performance Improvement Parameter Estimation Paired Data Nonnegative Matrix Factorization(NMF) Spoken Language Understanding Word Accuracy Wall Street Statistical Correlation Training Model Shared Task Critical Thinking Decoding Algorithm Bidirectional Lstm Automatic Translation Classification Task Domain-Specific Decision Fusion Embedded Device Elementary Students Gaussian Mixture Model(GMM) Hmm-Based Improving Accuracy Language Modeling Multiple Speakers Qualitative Characteristics Unlabeled Samples Recognition Rate Policy Gradient Method Two-Stage Assessment System Computer Assisted Language Learning Cross-Modality Embedding Model English Proficiency Foreign Language Fixed-Point Field Environment Hierarchical Topic Initialization Methods Learning Process Linear Regression Native Speakers Post-Processing Method Recognition Accuracy Unlabeled Data Training Method Speech Features Korean Speech Signal-To-Noise Training Phase Acceptance Or Rejection Automatic Transcription Augmentation Method Creative Thinking Fine-Tuning Gradient Descent Hidden Markov Model Learning Scenarios Optimal Weights Real-Time Random Selection Speech Recognition System External Knowledge Classification Accuracy Acoustic Characteristics Active Learning Method Auxiliary Networks Acoustic Model Training Count-Based Call Center Experimental Result I-Vector Learning System Loss Function Language Identification Linear Discriminant Analysis(LDA) Late Fusion Large-Scale N-Gram Lm Overlapped Speech Detection Output Prediction Proficiency Evaluation Proficiency Assessment Personalized Education Real-World Data Unsupervised Adaptation Speech Assessment Support Vectormachine(SVM) Speech Database New Method Learning Based Approach Supervised Learning Algorithm Accuracy Improvement Continuous Speech Recognition Comparative Study Dimension Reduction Education Research Ensemble Classifiers Early Fusion Entropy Loss Error Rate Far-Field Integrated System Multi-Layer Pseudo Labels Ami Corpus Benchmark Test Current State Feedforward Networks Government Agencies Lm Adaptation Performance Enhancement Rule-Based Approach Word Segmentation Training Results Training Cycle System Accuracy Small Set Reinforcement Learning Algorithm Audio-Visual Audio And Video Cross Entropy Confidence Measure Comparative Evaluation Document Classification Efficient Computation Empirical Study End To End(E2E) Language Model Adaptation Log Data Local Minima Layer Expansion Language Barrier Overall Performance Recognition Performance Text Data Text-Based Speech Feature Extraction Standard Deviation(STD) Time Factor