ETRI-Knowledge Sharing Plaform

Model Parameter Deep Neural Network(DNN) Endpoint Detection Statistical Model Computational Complexity Acoustic Features Word Error Rate Speech Signals CD-DNN-HMM Feature Vector Phone Recognition Non-Speech Speech Corpora Automatic Speech Recognition(ASR) Convolution Neural Network(CNN) Likelihood Ratio Test(LRT) Taylor Series Finite State Transducer Dnn-Based Acoustic Model Inter-Frame Performance Gain Front-End Model-Based Speech Recognition Accuracy Speech Corpus Recognition Performance Robust Speech Recognition Point Of Interest Two-Stage Signal-To-Noise Acoustic Source Automatic Translation Noise Reduction(NR) Cascaded Decision Semi-Supervised Training Time-Domain Non-Stationary Noise Signal Noise Ratio(SNR) Car Interior 3-Level Impact Noise Level Alignment Signal Detection Target Matrix Phone Error Rate Noise Separation Robot System Source Domain Temporal Dynamics Raw Data Unlabeled Data Polynomial Expansion Confidence Measure Human Knowledge Dynamic Features Voice Recognition Unsupervised Domain Adaptation Accurate Estimation Driving Conditions Spectral Characteristics Car Body Detection Error Microphone Arrays Computationally Efficient Mel-Frequency Cepstral Coefficient(MFCC) Real-Time Operation Signal Detector Scoring Model Reconstruction Method Computational Cost Double-Delta Decision Parameters External Knowledge Learning Methods Weight Matrix Test Set Unified Framework Road Condition Time-Scale Modification(TSM) Policy Gradient Method Isolated Word Spontaneous Speech Hidden Markov Model(HMM) Similarity Information Pre-Trained Model Nonlinear Function Error Back-Propagation Failure Rate Processing Method Dynamic Characteristics Speech Separation Service Performance Digital Signal Processing Objective Function Spectral Fluctuations Mechanical Sensing Car Navigation System Embedded Devices Speech Generation Prior Knowledge Decision Logic Wiener Filter Error Reduction Subjective Preference Estimation Method Target Location Factorization Approach Background Noise Channel Selection Linear Separability Robust Recognition High-Dimensional Features Resource-Constrained Localization Algorithm Retail Market Barrier-Free Filter Bank Sub-Frame Separation Device Training Cycle Spectrum Analyzer Speech Classification Level-Crossing Voice Command Parameter Optimization Data Augmentation Matrix Factorization Nonlinear Input Spatial Information Speech Emotion Recognition Self-Training Euclidean Norm Input Signal Manifold Regularized Reduction Technique Wall Street Cross-Correlation Function(CCF) Engineering Approach Frame-Level Nonlinear Activation Function Viterbi Algorithm Target Domain Probabilistic Decision Emotional Speech Hmm Based Speech Recognition Spectrum Reconstruction Synchronization Procedure Computational Auditory Scene Analysis Room Acoustic Unlabeled Samples Mobile Environment Conventional Methods Input Features Low-Resource Training Method Zero-Crossing Performance Degradation Acoustic Model Training Behavior Strategy Cost Function Generation Process Parameter Reduction Noise Suppression Sound Source Localization Blind Source Separation(BSS) Local Linear Nonlinear Manifold Feature Extraction Method Large Vocabulary Training Process Artificial Intelligence System Degraded Service Low Rank Matrix Vehicle Body Pseudo Labels Automatic Speech Recognizer Context-Dependent Human-Robot Interface Korean Speech Performance Improvement Parameter Space Cepstral Coefficients Handwritten Digits Handwritten Digit Recognition Linear Structure Moving Vehicle Data-Driven Learning Self-Improving Time-Frequency Mask Filter Estimation Speech Recordings Computational Efficiency Compensation Method Graph-Based Approach Knowledge-Based Conversational Speech Recognition Decision Knowledge Distributed Speech Recognition(DSR) Feedforward Neural Network(FNN) Language Barrier Normalized Cross-Correlation Speech Events Student Learning Reduction Rate Speech Quality Posterior Probability Estimation L1 And L2-Norm Manifold Structure Spectral Changes Speech Analysis Speech Recognition System Truncated Singular Value Decomposition Cross-Corpus Localization Method Post-Processing Graph Regularized Noisy Speech Synthesized Speech Back-Propagation Algorithm Reinforcement Learning Algorithm Statistical Hypothesis Testing Signal Distortion