Artificial bandwidth extension of speech signals using MMSE estimation based on a hidden Markov model.
Peter JaxPeter VaryPublished in: ICASSP (1) (2003)
Keyphrases
- hidden markov models
- speech signal
- speech recognition
- automatic speech recognition
- automatic speech recognition systems
- background noise
- noisy speech
- additive noise
- noisy environments
- linear prediction
- adaptive filtering
- conditional random fields
- speech enhancement
- kernel density
- speaker identification
- vocal tract
- hidden states
- speech synthesis
- acoustic features
- image segmentation
- sound source
- spectral subtraction
- word recognition
- hierarchical hidden markov model