MMSE-based stereo feature stochastic mapping for noise robust speech recognition.
Xiaodong CuiMohamed AfifyYuqing GaoPublished in: ICASSP (2008)
Keyphrases
- speech recognition
- noisy environments
- speech signal
- language model
- automatic speech recognition
- speaker identification
- background noise
- pattern recognition
- hidden markov models
- minimum mean square error
- speech processing
- speech recognition technology
- speech synthesis
- speech understanding
- computer vision
- cepstral coefficients
- speech recognizer
- digit recognition
- speech enhancement
- noisy speech
- keyword spotting
- speech recognition systems
- noise reduction
- speech recognition errors
- additive noise
- speaker recognition
- noise level
- signal to noise ratio
- expectation maximization
- denoising
- feature vectors
- feature extraction
- speaker independent
- information retrieval
- machine learning
- neural network