Noisy Constrained Maximum-Likelihood Linear Regression for Noise-Robust Speech Recognition.
D. K. KimM. J. F. GalesPublished in: IEEE Trans. Speech Audio Process. (2011)
Keyphrases
- noisy environments
- speech recognition
- linear regression
- maximum likelihood
- least squares
- speech signal
- noise reduction
- background noise
- speaker adaptation
- speech enhancement
- speech processing
- digit recognition
- automatic speech recognition
- speech recognizer
- speaker identification
- regression methods
- language model
- hidden markov models
- speech synthesis
- additive noise
- ridge regression
- nonlinear regression
- pattern recognition
- speech recognition technology
- word recognition
- gaussian distribution
- speech recognition systems
- signal to noise ratio
- speaker dependent
- information retrieval
- speech recognizers
- signal processing
- support vector machine
- training data
- computer vision