Keyphrases
Speech Recognition
90%
Hidden Markov Model
69%
Adaptation
40%
Language Model
32%
Language Modeling
32%
Sequential Learning
25%
Variational
25%
Bayesian Learning
25%
End-to-end Speech Recognition
24%
Recurrent Neural Network
21%
I-vector
19%
Neural Network
18%
International Cooperative Ataxia Rating Scale (ICARS)
17%
Hyperparameters
17%
Non-negative Matrix Factorization
17%
Speaker Adaptation
17%
Maximum a Posteriori
16%
Deep Neural Network
16%
Acoustic Modeling
15%
Speech Separation
14%
Speaker Recognition
14%
Latent Dirichlet Allocation
13%
Speaker Verification
13%
Source Separation
13%
Variational Autoencoder
12%
N-gram
12%
Sequence Data
12%
Latent Topics
12%
Linear Regression
12%
Variational Inference
12%
Transformer
12%
Latent Variable Models
12%
Domain Adaptation
11%
Telephone Speech
11%
Adversarial Learning
11%
Target Domain
11%
Temporal Convolutional Network
11%
Decoder
10%
Large Vocabulary Continuous Speech Recognition (LVCSR)
10%
Natural Language
10%
Posterior Collapse
10%
Encoder
10%
Long Short-term Memory
9%
Speech Signal
9%
Factor Analysis
9%
Variational Recurrent Autoencoder
9%
Speech Enhancement
9%
Predictive Classification
9%
Latent Variables
9%
N-gram Model
9%
Computer Science
Speech Recognition
100%
Language Modeling
61%
Neural Network
41%
Recurrent Neural Network
29%
Bayesian Learning
27%
Sequential Learning
26%
Experimental Result
21%
Source Separation
20%
Autoencoder
20%
Speaker Verification
20%
Adaptation Data
19%
nonnegative matrix factorization
19%
Leaning Parameter
18%
Latent Dirichlet Allocation
17%
Regularization
16%
maximum-likelihood
16%
Reinforcement Learning
15%
Speaker Recognition
14%
Independent Component Analysis
14%
Domain Adaptation
14%
Training Data
14%
Machine Learning
13%
Deep Neural Network
13%
Mutual Information
13%
Feature Extraction
13%
Linear Discriminant Analysis
12%
Dialog System
12%
Latent Variable Model
12%
Long Short-Term Memory Network
11%
Seq2Seq
11%
Attention (Machine Learning)
11%
Learning System
10%
Transformation Parameter
10%
And-States
10%
Learning Algorithm
10%
Blind Signal Separation
9%
Information Retrieval
9%
Recognition Performance
9%
Dirichlet Process
8%
Telephone
8%
Model Adaptation
8%
Marginal Likelihood
8%
Classification Error
7%
Classification Performance
7%
Unsupervised Learning
7%
Adversarial Machine Learning
7%
Semisupervised Learning
7%
Deep Learning Method
7%
Classification Accuracy
7%
Decision Trees
7%
Engineering
Recurrent Neural Network
27%
Model Parameter
26%
Gaussians
20%
Maximum a Posteriori
18%
Source Separation
18%
Matrix Factorization
17%
Experimental Result
16%
Reinforcement Learning
16%
Long Short-Term Memory
13%
Telephone
13%
Deep Neural Network
13%
Independent Component Analysis
12%
Regularization
12%
Robust Speech Recognition
11%
Basis Vector
11%
Speech Enhancement
11%
Dirichlet
10%
Autoencoder
9%
Maximum Likelihood
9%
Learning System
9%
Speech Signal
9%
Marginals
8%
Feature Extraction
8%
Deep Learning Method
8%
Single Channel
8%
Recurrent
8%
Conjugate Prior
7%
Latent Variable Model
7%
Posterior Probability
7%
Error Rate
7%
Recognizer
7%
Blind Signal Separation
7%
Signal-to-Noise Ratio
6%
Bayesian Approach
6%
Source Signal
6%
Acoustic Feature
6%
Joints (Structural Components)
6%
Mutual Information
6%
Bayesian Model
6%
Illustrates
6%
Continuous Speech Recognition
6%
Microphone Array
5%
Feature Vector
5%
Language Understanding
5%
Parameter Estimation
5%
Systems Performance
5%
Recursive
5%
Recognition Accuracy
5%
Hypothesis Test
5%
State Transition
5%
