A Training Method of Average Voice Model for HMM-Based Speech Synthesis.
Junichi YamagishiMasatsune TamuraTakashi MasukoKeiichi TokudaTakao KobayashiPublished in: IEICE Trans. Fundam. Electron. Commun. Comput. Sci. (2003)
Keyphrases
- cost function
- mathematical model
- objective function
- prior knowledge
- probabilistic model
- parameter estimation
- statistical model
- theoretical analysis
- similarity measure
- classification method
- input data
- modeling method
- test data
- prior information
- gaussian distribution
- evaluation method
- em algorithm
- bp neural network
- prediction model
- linear model
- training phase
- speech synthesis
- optimization method
- information retrieval
- energy function
- clustering method
- data mining
- reconstruction method
- probability distribution
- support vector machine
- high order
- linear regression
- tree structure
- detection method
- speech recognition
- training algorithm
- autoregressive
- bayesian framework
- neural network model
- feed forward neural networks