Improvements on Mel-Frequency Cepstrum Minimum-Mean-Square-Error Noise Suppressor for Robust Speech Recognition.
Dong YuLi DengJian WuYifan GongAlex AceroPublished in: ISCSLP (2008)
Keyphrases
- speech recognition
- noisy environments
- minimum mean square error
- cepstral coefficients
- signal to noise ratio
- speech signal
- hidden markov models
- noise reduction
- speech recognizer
- speaker identification
- automatic speech recognition
- pattern recognition
- language model
- speech synthesis
- speech recognition systems
- additive noise
- background noise
- low frequency
- speech enhancement
- recursive algorithm
- denoising algorithm
- noise level
- gaussian noise
- frequency domain
- edge detection
- information retrieval
- noise removal
- neural network
- mixture model
- feature set
- denoising
- image segmentation
- feature selection
- computer vision