LF-MMI Training of Bayesian and Gaussian Process Time Delay Neural Networks for Speech Recognition.
Shoukang HuXurong XieShansong LiuMax W. Y. LamJianwei YuXixin WuXunying LiuHelen MengPublished in: INTERSPEECH (2019)
Keyphrases
- speech recognition
- gaussian process
- gaussian processes
- automatic speech recognition
- isolated word
- marginal likelihood
- acoustic models
- gaussian process regression
- regression model
- bayesian framework
- expectation propagation
- model selection
- hidden markov models
- language model
- speech signal
- pattern recognition
- semi supervised
- latent variables
- hyperparameters
- speech recognizer
- posterior probability
- training set
- maximum likelihood
- speaker independent
- bayesian inference
- cross validation
- supervised learning
- neural network
- bayesian networks
- broadcast news
- face recognition
- error rate
- feature selection
- computer vision