Noise adaptive front-end normalization based on Vector Taylor Series for Deep Neural Networks in robust speech recognition.
Bo LiKhe Chai SimPublished in: ICASSP (2013)
Keyphrases
- speech recognition
- noisy environments
- taylor series
- neural network
- pattern recognition
- hidden markov models
- automatic speech recognition
- speech signal
- language model
- speaker identification
- speech recognition technology
- background noise
- speech synthesis
- speech recognition systems
- noisy speech
- noise reduction
- speech recognizer
- numerical integration
- speaker dependent
- noise model
- noise level
- natural images
- feature vectors
- computer vision