ScholarWorks@Hanyang University: Chang, Joon-Hyuk

Detailed Information

Chang, Joon-Hyuk (장준혁)

COLLEGE OF ENGINEERING (SCHOOL OF ELECTRONIC ENGINEERING)

Research Area: 음성음향신호처리 및 머신러닝

Tel 02-2220-0355
Homepage

ResearcherID ORCiD Scopus Google Scholar ID ResearchGate

Keyword

Voice activity detection:9||Deep neural network:9||deep learning:7||Blood pressure:7||Gaussian mixture model:7||Speech enhancement:7||weighted least squares:6||oscillometric blood pressure estimation:5||data augmentation:5||Speech synthesis:4||robust:4||Packet loss concealment:4||Localization:4||Speech recognition:4||semisupervised learning:4||bootstrap:4||sound event detection:4||Likelihood ratio test:3||deep neural network:3||ensemble:3||Adaptive muting:3||Sigmoid function:3||Signal processing algorithms:3||Location awareness:3||ITUT G.722:3||Ensemble:3||Convolution:3||Spectral gradient:3||Joint training:3||Statistical model:3||texttospeech:3||localization:3||Training:3||Convolutional neural network:3||Transformer:3||Deep learning:3||Task analysis:3||Kalman filters:3||Estimation:3||Confidence interval:3||automatic speech recognition:3||Embedded system:3||speech recognition:3||nonlineofsight:3||Convolutional Neural Network:2||Timeofarrival:2||thinfilm transistor (TFT):2||multiresolutional training:2||multispeaker:2||timeofarrival estimation:2||Feature enhancement:2||SMV:2||SVM:2||Feature extraction:2||Cache:2||Selectable Mode Vocoder (SMV):2||vital signal:2||voice activity detection:2||voice cloning:2||VoIP:2||Adaptive filter:2||noise suppression:2||CramerRao lower bound (CRLB):2||Blood pressure measurement:2||스펙트럼 변이:2||Noise classification:2||Weighted least squares:2||Computational complexity:2||Soft decision:2||Noise reduction:2||Noise suppression:2||Bandwidth extension:2||Computational modeling:2||Speaker verification:2||Conditional MAP:2||Gaussian Mixture Model (GMM):2||Spectral difference:2||adversarial training:2||Dualmicrophone:2||Gaussian mixture regression (GMR):2||radar:2||Maximum likelihood estimation:2||Global soft decision:2||representation learning:2||Decoding:2||Standards:2||Deep Learning:2||Prediction algorithms:2||Microphones:2||sensor fusion:2||Probability density function:2||Deep Neural Network:2||Support Vector Machine (SVM):2||confidence interval:2||continual learning:2||sigmoid:2||Support vector machine:2||skipped filter:2||sleep stage:2||Automatic speech recognition:2||soft decision:2||sound source localization:2||convolutional recurrent neural network:2||speaker adaptation:2||speaker verification:2||machine learning:2||curriculum learning:2||Deep neural networks:2||Textindependent speaker verification:2||maximum likelihood estimation:2||Reverberation:2||mean square error methods:2||mean square error performance:2||medical device:2||microphone:2||deep neural networks:2||Machine learning:1||Mahalanobis distance:1||Mathematical models:1||Measurement uncertainty:1||Median filter:1||MelFrequency Cepstral Coefficients (MFCC):1||MelGAN:1||Memory:1||Microphone array nonlinear acoustic echo cancellation:1||Minimum bias:1||Minimum classification error:1||Minimum mean squared error:1||Minimum statistics:1||Modulated complex lapped transform (MCLT):1||Monte Carlo methods:1||Monte Carlo simulation:1||Multi speaker speech synthesis:1||Multichannel linear prediction:1||Multimicrophone nearend speech covariance:1||Multispeaker:1||Multitap least squares estimator:1||MultiSpeech:1||Multiple linear regression:1||Multiple linear regression analysis:1||Multiple regression:1||Multipleinstance learning:1||Nanowires:1||Neural network:1||Noise Estimation:1||Noise awareness:1||Noise measurement:1||Noise power estimation:1||Nonlineofsight (NLOS):1||Optimization:1||Oscillometric blood pressure measurement:1||Oscillometric measurement:1||Oscillometric method:1||PMWF:1||ParaNet:1||Parameterized multimicrophone Wiener filter:1||Parametric bootstrap:1||Parametric multichannel Wiener filter:1||Parametric shaping function:1||Photoluminescence:1||Pollution measurement:1||Polynomial regression:1||Position estimation:1||Prior distribution:1||Raisedcosine function:1||Raman:1||Reference locality:1||Reflection:1||Regression:1||Reinforcement learning:1||Respiration rate:1||Reverberant speech recognition:1||Reverberation time:1||Robust:1||Robust speech recognition:1||Robust statistics:1||Robustness:1||Sample mean:1||Sample median:1||Secondorder Conditional Maximum a posteriori (Secondorder CMAP):1||Selectable mode vocoder:1||Sensors:1||Sequence discriminative training:1||Sequence to sequence:1||Shrinkage factor:1||Sigmoid:1||Smartphone:1||SnO2/MgO:1||Sound source localization:1||Source localization:1||Spacealternating generalized:1||Sparse Feature:1||Spectral Difference:1||Spectral coherence:1||Spectrogram:1||Speech Recognition:1||Speech absence probability:1||Speech dereverberation:1||Speech processingReverberationMicrophonesNeural networksFeature extractionSpeech recognitionPrediction algorithmsDeep neural networkoffline processingsingle microphonespeech dereverberationweighted prediction error:1||Speech/Muisc Classification Algorithm:1||Speech/Music Classification Algorithm:1||Speech/Music classification algorithm:1||Speech/music classification:1||StahelDonoho estimator:1||Statespace modeling:1||Statistical Model:1||Statistical testing:1||Steepest descent:1||Stein unbiased risk estimator:1||Stereo microphones:1||Support vector machine (SVM):1||TDOA:1||TSNR:1||Tacotron2:1||Taylor series approximation:1||Television audience measurement:1||Testing:1||Texttospeech:1||Time domain speech enhancement:1||Timedomain analysis:1||Timefrequency analysis:1||Training data:1||Transmission electron microscopy (TEM):1||VSSSSAF:1||Value of q:1||Variable step size:1||Variable stepsize:1||Vehicle Noise:1||Vital signals:1||Voice Activity Detection:1||Voice cloning:1||Volterra filter:1||WaveUNet:1||Weakly supervised learning:1||Weighted finite state transducer:1||Weighted prediction error:1||Weighting matrix:1||Wiener Filter:1||Wireless communication:1||a":1||acoustic beamforming:1||acoustic echo suppression:1||acoustic model:1||acoustic scene classification:1||active impulsive noise control:1||active noise control:1||adaptive filter:1||adaptive filtering:1||adaptive filters:1||adaptive kernel:1||alphaTrimmed mean:1||antispoofing:1||artificial intelligence:1||attention:1||attention alignment:1||attention masking:1||attention mechanism:1||attentionbased endtoend speech synthesis:1||attentive statistics pooling:1||audio scene classification:1||banddependent variable stepsize:1||beamforming:1||block implementation:1||blood pressure:1||bootstrapinspired technique:1||bottleneck layer:1||bottleneck network:1||class activation maps:1||classification algorithm:1||clustering:1||compensation circuit:1||convex programming:1||convolutional neural network:1||convolutional neural network (CNN):1||convolutional neural networks:1||cost function:1||crosslingual:1||crossmodal distillation:1||crosstower:1||datadriven approach:1||deep Boltzman machine:1||deep neural network (DNN):1||deep neural network calibration:1||deep neural networks (DNNs):1||deep supervision:1||dereverberation:1||diffusion model:1||directionofarrival:1||discriminative weight training:1||distant speech recognition:1||distributed information systems:1||documentlevel neural TTS:1||domain adaptation:1||dualmicrophone:1||electrocardiography (ECG):1||electronic manometer for home:1||embedded software:1||embedded systems:1||endofturn detection:1||endofutterance (EOU) detection:1||endtoend:1||endtoend neural diarization:1||endtoend speech recognition:1||enhancement:1||estimate:1||estimation:1||estimation theory:1||existing measurementbased method:1||expectationmaximization (SAGE):1||extended Kalman filter:1||extrapolated single propagation unscented Kalman filter:1||feature embedding:1||featurewise linear modulation:1||filtering theory:1||finetuning:1||formant:1||formant enhancement:1||gaussian mixture model:1||global soft decision:1||global style tokens:1||guided score:1||heterojunction:1||humanrobot interaction:1||hybrid deep neural network:1||hybrid system:1||hypertangent loss function:1||identification:1||impulse noise:1||impulsive interferences:1||indium gallium tin oxide:1||individualweightingfactors sign subband adaptive filter:1||indoor navigation:1||information theoretic learning:1||infrared sensor:1||iterative algorithms:1||joint optimization:1||joint training:1||kmeans clustering:1||kmedoids clustering algorithm:1||knearest neighbor:1||kalman filter:1||knowledge distillation:1||l(1)norm minimization:1||language model:1||language model (LM):1||latent feature:1||least median of squares:1||likelihood ratio test:1||lineofsight:1||linear predictive coding coefficients:1||location awareness:1||location parameter:1||logpower spectra:1||lowpressure cooled exhaust gas recirculation:1||luminance degradation:1||masked frequency modeling:1||masking:1||maximum a posteriori:1||maximum amplitude algorithm (MAA):1||maximum likelihood (ML):1||maximum likelihoodtype estimator (M estimator):1||mean square error:1||mercury sphygmomanometer:1||minimisation:1||minimum classification error:1||mobile communication:1||multichannel speech extraction:1||multistage maximum likelihoodtype (MM) estimator:1||multistage maximum likelihoodtype estimator (MM estimator):1||multitap least squares estimator:1||multitask learning:1||multipleinput multipleoutput (MIMO):1||multivariate Gaussian distribution:1||network speech recognition:1||neural architecture search:1||neural beamformer:1||noise classification:1||noise estimation:1||noise reduction:1||noiserobust classification:1||nondifferentiable convex problem:1||nonlinear acoustic echo suppression:1||nonlinear clipping detector:1||nonlinear distortion:1||nonlinear echo:1||nonparametric bootstrap:1||offline processing:1||online speech recognition:1||organic lightemitting diode (OLED):1||orthogonal series:1||oscillometry blood pressure estimation:1||packet loss concealment (PLC):1||parameter estimation:1||parameter sharing:1||pause hesitation:1||penalty function:1||phase shift keying:1||photoplethysmography (PPG):1||phototransistor:1||position estimation:1||power level difference ratio:1||pretraining:1||previous information:1||prior location:1||probability density function:1||projections:1||prosody:1||q값:1||radio navigation:1||random forest:1||range measurements:1||received signal magnitude:1||regression:1||regression model:1||residual gain estimation:1||robust TOA source localisation algorithm:1||robust range estimation algorithm:1||robust signal processing:1||robust speech recognition:1||robustness:1||secondorder conditional MAP:1||selfattention mechanism:1||selfdistillation:1||sensor placement:1||sequential deep neural network:1||shortduration speaker verification:1||shrinkage factor:1||shrinkagebased sequential source localisation and range estimation algorithm:1||signal processing:1||signaltonoise ratio (SNR):1||single microphone:1||smart wristwatch:1||sparse Volterra systems:1||spatial sign:1||speaker diarization:1||spectral difference:1||spectral gradient:1||speech dereverberation:1||speech emotion recognition:1||speech endpoint detection (EPD):1||speech enhancement:1||speech processing:1||speech quality:1||speech separation:1||speechpresence uncertainty:1||speech/music classification algorithm:1||spherical simplex unscented transform:1||spoken dialogue system:1||spoofingaware speaker verification:1||squared error function:1||statistical analysis:1||statistical modelbased speech enhancement:1||statistical testing:1||streaming:1||tellurium oxide:1||temporal convolutional network:1||time delay:1||time series forecasting model:1||timedomain speech separation:1||timeofarrival source localisation algorithms:1||turbocharged gasoline direct injection engines:1||twostep:1||variable stepsize sign subband adaptive filter:1||variablelength:1||variational Bayes:1||variational Bayesian:1||versoria function:1||virtual acoustic channel expansion:1||voiced/unvoiced classification:1||wearable wristwatch:1||weighted prediction error:1||가중 유한 상태 전이기:1||딥러닝:1||분류:1||상황 인지:1||서포트벡터머신:1||시퀀스 분류 학습:1||어텐션:1||음성 검출기:1||음성/음악 분류 알고리즘:1||음성인식:1||음향 강화:1||임베디드 시스템:1||조건 사후 최대 확률:1||차량 소음:1||통계적 모델:1||트랜스포머:1||합성곱 신경망:1||":1||회귀:1||(1)norm minimisationbased variable stepsize:1||A priori SNR:1||A priori SNR estimation:1||Acous tic environment classification:1||Acoustic Model:1||Acoustic beamforming:1||Acoustic data transmission system:1||Acoustic echo cancellation:1||Acoustic echo canceller:1||Acoustic echo suppression:1||Acoustic echo suppressor:1||Acoustic environment classification:1||Acoustic model (AM):1||Acoustic scene classification:1||Acoustics:1||Adaptive filters:1||Adaptive multirate wideband:1||Adaptive selection:1||Affine projection sign algorithm:1||Artificial noise:1||Attention:1||Attention based latent feature:1||Attention mechanism:1||Audio matching:1||Automated machine learning (AutoML):1||Auxiliary encoder:1||Bayesian inference:1||Best linear unbiased estimator:1||Biased estimation:1||Blind estimation:1||Block Newton method:1||Blood pressure (BP):1||Blood pressure (BP) estimation:1||Bootstrap:1||Bootstrap technique:1||Breakdown point:1||CRLB estimation:1||Classification:1||Codecs:1||Computer architecture:1||Conditional maximum a posteriori:1||Continuous wave radar:1||Convergence:1||Convolutional neural networks:1||Covariance matrices:1||CramerRao lower bound estimation:1||DFT:1||Data models:1||Datadriven training method:1||Decay rate:1||Decay rate variance:1||Decisiondirected method:1||Deep Belief Network:1||Deep Qnetwork:1||Deep speaker embedding:1||DeepVoice3:1||Delayless structure:1||DempsterShafer fusion:1||DempsterShafer theory:1||Dereverberation:1||Deviation:1||Digital signal processing:1||Direction of arrival:1||Directionofarrival estimation:1||Discriminative weight training:1||Drone:1||E2E speech recognition:1||Eigenvalue decomposition:1||Electrocardiography:1||Embedded System:1||End to end automatic speech recognition:1||Ensemble Model:1||Ensemble acoustic model:1||Error variance:1||Expectation maximization:1||Expectation maximization (EM):1||Explicit spatial filtering:1||Feature Extraction:1||FiLM conditioning:1||Filtering:1||Filtering techniques:1||Fingerconducted speech transmission:1||First peak:1||Fusion:1||GMMbased clustering:1||Gain:1||Gaussian mixture model (GMM):1||Gaussian mixture regression:1||Global style token:1||Gunfire Localization, Sound Localization:1||HMM based hybrid ASR:1||Hampel filter:1||Hidden Markov model:1||HodgesLehmann estimator:1||Impulsive noise:1||Incylinder pressure:1||Independent vector analysis:1||Indexes:1||Influence function:1||Intelligent sensors:1||Interference:1||Iterative decoding:1||Joint training framework:1||Kaldi:1||Kalman filter:1||LedoitWolf method:1||Lexicon:1||Licenses:1||Linguistics:1||Loss function:1||Lowrank approximation:1||M estimator:1||ML estimation:1||MM estimator:1||MSE performance:1||

RSS_1.0 RSS_2.0 ATOM_1.0

222, Wangsimni-ro, Seongdong-gu, Seoul, 04763, Korea+82-2-2220-1366

Certain data included herein are derived from the © Web of Science of Clarivate Analytics. All rights reserved.
You may not copy or re-distribute this material in whole or in part without the prior written consent of Clarivate Analytics.

Detailed Information

Keyword

BROWSE