Decomposition of speech into voiced and unvoiced components based on a state-space signal model.
Mark ThomsonSimon BolandMike WuJulien EppsMichael SmithersPublished in: ICASSP (1) (2003)
Keyphrases
- state space
- probabilistic model
- mathematical model
- objective function
- computational model
- image decomposition
- measured data
- conceptual model
- experimental data
- markov random field
- cost function
- frequency domain
- speech recognition
- prior knowledge
- state variables
- speech signal
- analytical model
- bayesian networks
- high level