Speech Enhancement for Automatic Speech Recognition Using Complex Gaussian Mixture Priors for Noise and Speech.
Ramón Fernandez AstudilloEugen HoffmannPhilipp MandelartzReinhold OrglmeisterPublished in: NOLISP (2009)
Keyphrases
- speech enhancement
- noisy environments
- noise reduction
- speech signal
- signal to noise ratio
- noisy speech
- gaussian mixture
- single channel
- speech recognition
- vocal tract
- background noise
- linear prediction
- multi channel
- wiener filter
- smoothing algorithm
- edge preserving
- automatic speech recognition
- additive noise
- speech synthesis
- prior information
- pattern recognition
- density estimation
- closed form
- edge detection
- expectation maximization
- information retrieval
- gaussian mixture model
- probability density function
- bayesian framework
- image processing