Multi-microphone speech dereverberation using expectation-maximization and Kalman smoothing.
Boaz SchwartzSharon GannotEmanuël A. P. HabetsPublished in: EUSIPCO (2013)
Keyphrases
- expectation maximization
- em algorithm
- automatic speech recognition
- speech recognition
- speaker diarization
- gaussian mixture model
- maximum likelihood
- maximum likelihood estimation
- speech signal
- mixture model
- image smoothing
- filtering algorithm
- audio visual
- broadcast news
- speaker identification
- speaker recognition
- kalman filtering
- speech synthesis
- visual information
- motion parameters
- maximum a posteriori
- kalman filter
- image segmentation
- image restoration
- human computer interaction
- low level
- hidden markov models
- feature vectors
- multiscale
- bayesian networks
- image sequences