Incorporating a Generative Front-End Layer to Deep Neural Network for Noise Robust Automatic Speech Recognition.
Souvik KunduKhe Chai SimMark J. F. GalesPublished in: INTERSPEECH (2016)
Keyphrases
- automatic speech recognition
- noisy environments
- neural network
- speech recognition
- speech signal
- discriminative training
- multi layer
- noise reduction
- hidden markov models
- word error rate
- broadcast news
- spoken words
- speech retrieval
- pattern recognition
- word recognition
- generative model
- artificial neural networks
- signal to noise ratio
- single layer
- acoustic features
- speech enhancement
- speaker adaptation
- speech corpus
- multiscale
- image processing