Discriminative learning of additive noise and channel distortions for robust speech recognition.
Jiqing HanMunsung HanGyu-Bong ParkJeongue ParkWen GaoDoosung HwangPublished in: ICASSP (1998)
Keyphrases
- speech recognition
- additive noise
- speech signal
- discriminative learning
- noisy environments
- hidden markov models
- automatic speech recognition
- noisy images
- generative model
- language model
- highly efficient
- maximum likelihood
- spatial domain
- speech enhancement
- random variables
- pattern recognition
- neural network
- image patches
- frequency domain
- graphical models
- video sequences
- computer vision
- machine learning