Speech Enhancement Based on Speech Spectral Complex Gaussian Mixture Model.
Guo-Hong DingXia WangYang CaoFeng DingYuezhong TangPublished in: ICASSP (1) (2005)
Keyphrases
- gaussian mixture model
- speech enhancement
- speaker recognition
- speaker identification
- speech signal
- linear prediction
- noisy environments
- vocal tract
- mixture model
- noise reduction
- signal to noise ratio
- single channel
- em algorithm
- spectral analysis
- feature vectors
- expectation maximization
- speech recognition
- maximum likelihood
- spectral features
- image coding
- hidden markov models
- noisy speech
- automatic speech recognition
- background subtraction
- smoothing algorithm
- prior information
- additive noise
- background noise
- image compression
- feature space