ETRI-Knowledge Sharing Plaform

Deep Neural Network(DNN) Network-Based Mpeg Surround Interactive Music Speech Coding Audio Features High Performance Proposed Model Subjective Listening Tests Audio Codec Music Application Emotional Speech Spatial Cue Speech Synthesis System Post-Processing Video Service Sound Source Sound Quality Speech Codec Superior Performance Conventional Methods Analysis And Synthesis Side Information Non-Speech File Format Full-Band Generative Models Low Bit Rate Bit Allocation Text-To-Speech Synthesis Audio Analysis Digital Music High-Quality Deep Learning(DL) Music Industry Stereo Audio Audio Synthesis Generative Adversarial Network Latent Space Multichannel Audio Convolutive Source Separation Statistical Parametric Speech Synthesis(SPSS) Additional Information Synthesized Speech Speech Signals Perceptual Quality Audio Description Learning-Based Acoustic Parameters Subjective Evaluation Head Related Transfer Function(HRTF) Speech Detection Novel Method Information Reconstruction Acoustic Echo Cancellation Audio Information Audio Data Audio Signal Processing Audio Feature Extraction Coding Gain Discrimination Method Dynamic Characteristics Different Dimensions Data Compression Distortion Ratio Flexible Features Filtering Method Low-Resource Level Difference Moving Picture Experts Group(MPEG) Media Content Modified Discrete Cosine Transform(MDCT) Perceptual Similarity Spectral Parameters Partitioned Block Frequency Domain Prediction And Analysis Parameter Selection Virtual Source Signal Characteristics Source Location Separation Algorithm Separation Performance Weighted Sum Text-To-Speech System Time-Frequency Resolution High-Dimensional Adversarial Training Bandwidth Extension Coupling Effects Control Algorithm Convolution Neural Network(CNN) Enhanced Structure Frame Rate Independent Control Location Information(GPS) Low-Rate Root Mean Square(RMS) Sub-Band Source Characteristics Speech Discrimination Power Spectra Waveform Synthesis Second Order Statistics(SOS) Network Architecture Noise Reduction(NR) Objective Evaluation Objective Assessment Training Data Reconstruction Quality Blind Source Separation(BSS) Backward Compatibility Convergence Speed De-Quantization Diffusion Model Entropy Model Emotion Classification Emotional Expression Feature Transformation Gradient Method High-Pass Filtering Model Compression State-Of-The-Art Selection Strategy Source-Filter Model Semi-Automatic Speech Features Pitch-Synchronous Warped Linear Prediction Open Source Signal Generation Temporal Structure Automatic Music Transcription(AMT) Audio Quality As 4 Basic Media Convolutive Mixtures Control Method Coding Scheme Compression Efficiency Excitation Signal Frequency Dependent Fine-Tuning Glottal Closure Instants Linear Prediction Analysis Low-Dimensional Multichannel Signals Memory Usage Skip Connections Scalar Quantization Speech Enhancement Synchronous Approach Speech Communication Psychoacoustic Model Pitch Information Perceptual Weighting Vector Space Various Approaches Post-Processor Neutral Speech Reconstruction Performance Residual Error Acoustic Characteristics Backward Compatible Blind Deconvolution Computational Resources Classification Accuracy Coding System Design Scheme Discrete-Time Effective Frequency Fixed Dimension Feature Extraction Method Flow Factor High Efficiency Joint Diagonalization Multi-Band Numerical Experiments Relative Gradient Speaker Embeddings Selection Problem Separation Techniques Speech Analysis Synthesis Method Sinusoidal Vibration Permutation Ambiguity Perceptual Evaluation Of Speech Quality Vocal Tract Visualization Tool Waveform Reconstruction Weight Control Weight-Based Speech Separation Training Criteria Transfer Function Model Audio Recording Computational Efficiency Error Rate End-To-End Learning Emotion Control Implementation Scheme Model Complexity Siamese Network Sub-Nyquist Sampling Stereo Coding Perceptual Speech Quality Spatial Image Side Effects Spatial Audio Coding Stereo Sound Telepresence System Nyquist Rate Nonlinear Function Optimal Transformation Optimization Techniques(OT) Time-Domain Time-Frequency Characteristics Representation Learning Audio Signal Compression Based Approach Cbs Algorithm Control Technique Coding Models Data Augmentation Enhancement Scheme End-To-End Training Frequency Domain Adaptive Filter Frequency-Domain Method Hearing Impaired People High-Band Multi-Stage Vector Quantization Narrow Bandwidth Speech Characteristics Spectral Bandwidth Subjective Quality Musical Rhythm Subjective Test User Interactivity Vibration Signal Subjective Assessment Tactile Device Objective Test Training Samples Audio Compression Analysis-By-Synthesis Blind Separation Compact Representation Continuous Space Coding Algorithm Fundamental Frequency Independent Component Analysis Interpolation Technique Low Pass Filter Music Player Mobile Station(MS) Noisy Data Signal Reconstruction Signal Processing Techniques Speaker Recognition Simulation Results Variable Step Size Sensor Noise